GeolCarp_Vol70_No2_91

GEOLOGICA CARPATHICA

, APRIL 2019, 70, 2, 91–112

doi: 10.2478/geoca-2019-0006

www.geologicacarpathica.com

The Plio–Pleistocene demise of the East Carpathian

foreland fluvial system and arrival

of the paleo-Danube to the Black Sea

ANDREI MATOSHKO



, ANTON MATOSHKO

and ARJAN DE LEEUW

3, 4

GeoExpert LLc., freelance geologist, Ap. 37, 1/25 Draizera Str., 02217, Kyiv, Ukraine;



andreimatoshko@gmail.com

GeoExpert LLc., freelance geologist, Ap. 37, 1/25 Draizera Str., 02217, Kyiv, Ukraine; uxz357@gmail.com

Université Grenoble Alpes, Institut des Sciences de la Terre (ISTerre), CS40700, 38058 Grenoble, France;

arjan.de-leeuw@univ-grenoble-alpes.fr

Formerly at CASP, West Building, Madingley Rise, Madingley Road, Cambridge, CB3 0UD, United Kingdom

(Manuscript received July 31, 2018; accepted in revised form February 18, 2019)

Abstract: This paper studies the Porat Formation (Fm.), which was deposited along the NE margin of the Dacian Basin

part of the East Carpathian foreland (ECF) during the Pliocene and Early Pleistocene. We use a review of stratigraphic

data in combination with lithofacies and sedimentary architecture analysis to interpret the Porat Fm. as a large sandy

alluvial basin infill with an aggradational structure, consisting of cyclic successions of shallow sandy high-energy braided

rivers. Aggradation of the Porat Fan was governed by subsidence of the Dacian Basin, along with a northerly supply of

water and sediment from the Carpathians. Along the southern margin of the area the fan entered the Reni–Izmail-Trough,

which formed the periodically active gateway between the Black Sea and Dacian basins. Along this trough, the Porat Fm.

is developed in a different facies, discerned as the Dolynske Member (Mb. 1), which accumulated in the channel of a large

river interpreted as the paleo-Danube. According to mammal stratigraphy of the Porat Fm. this continental-scale river had

reached the area by the Gelasian to early Calabrian. The Porat alluvial infill indicates a stable water supply from

the Carpathians, which explains the ecologically mixed fauna in its deposits: moistened forested alluvial plain-valleys

were present between the zonally semi-arid steppe interfluves. The Porat Fm. and the previously studied late Miocene

Balta Fm. are key elements for further in-depth study of the terrestrial evolution (tectonic–sedimentary–relief) of the ECF

and north-western Black Sea coastal regions.

Keywords: Paratethys, East Carpathian foreland, Pliocene, Quaternary, alluvial deposits.

Introduction

This paper sheds light on the geology of the fluvial Porat

Formation (hereafter “Fm.”), which accumulated in the border

region of Romania, Moldova and Ukraine during the Pliocene–

Gelasian–Calabrian (Fig. 1). This area (hereafter “Porat area”)

occupied a key position between the Black Sea and Dacian

basins, which were part of the enormous almost land-locked

Paratethys Sea (Rögl 1998; Popov et al. 2004). As it is estab-

lished in the course of the present study, the south-eastern

corner of the Porat area, a trough along the northern margin of

the Dobrogea Orogen, sometimes served as a connecting

straight between the Black Sea and Dacian basins and at other

times it formed a terrestrial barrier. Despite the high level of

outcropping, abundance of fossil material and great number of

local publications, there is no modern synthesis of the geolo-

gical development of the Porat area. There are currently a num-

ber of competing paleogeographic and sedimentary concepts

and local stratigraphic schemes with multiple names for

the same stratigraphic units in different locations.

Our research objective was to distinguish and map fluvial

formations and their elements for the subsequent reconstruc-

tion of the fluvial sedimentation and relief evolution of the study

area, in tight correlation with tectonic, climatic and sea-level

events. We explored the geology of the Porat area, with a par-

ticular emphasis on its rich record of fluvial deposits, most

informative among its analogues present along the north-

western coastal area of the Black Sea (Matoshko et al. 2009).

In this paper we also make an effort to combine and juxtapose

the different datasets, approaches and views of various groups

of researchers from Romania, Moldova, Ukraine, Russia,

Netherlands as well as from some other countries. In our work,

we have paid particular attention to the role of tectonic move-

ments as a factor of influence on the reconstructed sedimen-

tary processes.

How large parts of the brackish–marine sedimentary envi-

ronment of the Paratethys were converted into new land

remains a picture painted by rough strokes. Nevertheless,

the last study (Matoshko et al. 2016) confirmed that fluvial

formations are very informative objects for the comprehension

of sedimentation and morphogenesis in the vicinity of the land-

sea borderline during basin retreat. During the late Miocene,

a large-scale fluvio–deltaic system known as the Balta For-

mation (Matoshko et al. 2016) developed in the East Carpathian

foreland (hereafter ECF). Its progressive progradation was

interrupted by a marked transgression at the base of the Pontian

MATOSHKO, MATOSHKO and DE LEEUW

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

regional stage (corresponding to the late Miocene, Messinian).

Our current study addresses the subsequent, Porat Stage

(Pliocene–Gelasian–Calabrian), for which we discern Early

and Late Porat substages in the evolution of the ECF, when

the previously widespread Balta system was transformed and

alluvial deposits started to accumulate in a number of more

localized depocenters, among which the Porat area. This can

be considered a transitional stage towards the eventual for-

mation of the familiar cut-and-fill terraced river valleys of

the Pleistocene. We furthermore lay an accent on the evolution

of the NE margin of the Dacian Basin and its connection

directly to the Black Sea during the Pliocene–Gelasian–

Calabrian through a detailed study of the fluvial Porat Fm.

Another important question which we address is the age of

the first appearance of the Danube in this critical gateway area

between the Dacian Basin and the Black Sea.

Estimates for the timing of the arrival of the continental-

scale Danube River to the Black Sea are debated and range

from the late Miocene (Clauzon et al. 2005) to Pleistocene

(Wong et al. 1994). Clauzon et al. (2005) suggested that

the Danube arose during the supposed Messinian desiccation

of the Dacian Basin and the Black Sea. Andreescu (2009),

on the contrary, inferred that the river pierced through

the Iron Gates at the western margin of the Dacian Basin about

2.0–1.8 Ma ago, based on an influx of fresh-water molluscs

from the Pannonian Basin as well as alluvial deposits in

Fig. 1. Location map and map of outcrops used for the study. Names of tectonic units and faults according to Matenco et al. (2007): EEP — East

European Platform, SP — Scythian Platform, ND — North Dobrogea, MP — Moesian Platform, FD — Focşani Depression, BF — Bistrița

Fault, NTF — New Trotus Fault, PCF — Peceneaga–Camena Fault; according to Bala et al. (2003): SGF — St. Gheorghe Fault. Outcrop names

linking to numbers:1 — Măluşteni 1; 2 — Tuțcani; 3 — Mȃnzătești; 4 — Bereşti; 5 — BăIăbăneşti 1; 6 — BăIăbăneşti 2; 7 — Slobodzia

Oancea; 8 — Foltești; 9 — Ljidileni 1, 2; 10 — Tuluceşti; 11 — Vanatori; 12 – Cociulia 1; 13 — Lărguța; 14 — Crăciun; 15 — Ciobalaccia 1;

16 — Ciobalaccia 2; 17 — Chioselia; 18 — Cîșla; 19 — Flocoasa; 20 — Chioselia Mare; 21 — Frumușica; 22 — Holuboaia-Doina;

23 — Borceag; 24 — Spicoasa; 25 — Tătăreşti 1; 26 — Tătăreşti 2; 27 — Andrușul de Sus; 28 — Luceşti; 29 — Tartaul de Salcie;

30 — Trifestii Noi; 31 — Albota de Jos; 32 — Hîrtop-Balabanu; 33 — Moscovei; 34 — Cahul 2; 35 — Dermengi; 36 — Budăi 1; 37 — Ursoaia;

38 — Budăi 2; 39 — Pelinei; 40 — Musaitu; 41 — Manta; 42 — Vladimirovca; 43 — Gavanoasa; 44 — Colibași 1; 45 — Vynogradivka 1;

46 — Colibași 2; 47 — Vulcănești 1; 48 — Colibași-Brînza; 49 — Bolgrad 1; 50 — Vulcănești 2; 51 — Vulcănești 3; 52 — Bolgrad 3;

53 — Valeni 1; 54 — Valeni 2; 55 — Topolyne; 56 – Valeni–Slobodzia Mare 1; 57 — Valeni–Slobodzia Mare 2; 58 — Valeni–Slobodzia Mare

3, 4; 59 — Cişmichioi 1, 2; 60 — Cislita–Prut; 61 — Etulia Noua 6; 62 — Etulia Noua 2; 63 — Kotlovyna; 64 — Giurgiuleşti 1;

65 — Dolynske 1; 66 — Dolynske 2; 67 — Giurgiuleşti 2 (Ripa Scortselskaia); 68 — Dolynske 3; 69 — Dolynske 4; 70 — Lymanske 1

(see GPS coordinates of the outcrops in Appendix). Inset map: E. — Eastern, S. — Southern, B.Z. — Bend Zone.

THE PLIO–PLEISTOCENE FLUVIAL SYSTEM OF THE PALEO-DANUBE

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

the western part of the Dacian Basin, meaning its arrival to

the Black Sea could only have occurred later. According to

Olariu et al. (2018), the Danube appeared on the Romanian

Black Sea shelf at about 4 Ma, as indicated by an increase in

sediment volume above an intra-Dacian erosional surface and

a coincident change in the architectural style in the adjacent

deepwater fan. The only available dated provenance record

from the Black Sea Basin (de Leeuw et al. 2018), indicates

that Danube-supplied sediment first arrived sometime between

4 Ma and 1 Ma ago, but does not provide more conclusive age

constraints either. A factor that has remained completely

under-addressed in in this discussion is the geology of the final

gateway of the Danube to the Black Sea, which we address in

detail here.

Geological setting

Designation of the Porat Fm. and Dolynske Mb.

The Porat area has a long research history, albeit mostly

restricted to publications in Romanian, Ukrainian and Russian.

We here build on some recent reviews (Gozhik 2006; Mokriak

et al. 2008; Matoshko et al. 2009) that have renewed and

restructured the published information on our study area to

some extent. Scattered data regarding the Pliocene–Quaternary

faunal remains unearthed in this area were first reported during

the 19

century. Much later Konstantinova (1967 with refe-

rence to Pavlov 1925) suggested naming the sands with gravel

and Roussillon fauna near the Prut River the Porat Fm., which

referred to the Roman name for the Prut River: “Porata”.

The formation was further divided into Lower and Upper

Porat units.

Meanwhile, Ghenea (1968, 1997) carried out research

within the northern part of the Bârlad–Prut interfluve in

Romania (Fig. 1), where he encountered deposits with a very

similar nature to the Porat deposits on the Moldavian side of

the Prut. He described the terrestrial deposits of the Bereşti–

Măluşteni Fm. with an Astian large mammal fauna and the

Tuluceşti Fm., which has a Villafranchian fauna. These con-

cepts were not modified significantly hitherto, although

the sub sequent introduction of the BăIăbăneşti Fm., which

supposedly overlaps the Bereşti–Măluşteni Fm. and may be

correlated with the Tuluceşti Fm. (Ionesi et al. 2005; Enciu &

Dumitricǎ 2008), complicated matters slightly. On the Roma-

nian state geological maps (Ghenea & Ghenea 1967; Saulea et

al. 1967) the Bereşti–Măluşteni Fm. was absorbed into

the depicted Levantine, whereas the Tuluceşti and BăIăbăneşti

formations were attributed to the Early Pleistocene.

Four complexes of alluvial and alluvial–deltaic deposits

were distinguished by Gozhik & Chirca (1973) based on col-

lections of freshwater molluscs from borehole cores taken

between the Danube, Cahul and Yalpukh lakes (Fig. 1). They

attributed these fluvial complexes to the Cimmerian,

Kuialnykian and Gurian stages (Middle Zanclean–Calabrian,

Fig. 2).

Synthetic sections derived from a number of outcrops along

the Bolshaya Salcia and Cahul rivers, as well as two boreholes

led Hubca (1982) separate the so-called Carbalia Beds from

the Porat Fm. The only argument for separation was a slight

difference in fossil assemblages.

Over the course of the Moldavian state geo-environmental

survey accomplished in 1994, the Prut-Yalpukh interfluve in

Moldova was geologically mapped in detail. The resulting

maps depict the Musaitu and Cîşlița-Prut suites that on

the accom panying geological profiles clearly match the Lower

Porat unit as defined before by Roşca (1969).

We have here chosen to absorb the numerous Romanian,

Moldavian and Ukrainian units (most of them described

above) into the Porat Fm. based on their lithological and strati-

graphic similarity. On the other hand, we have decided to dis-

tinguish the Dolynske Member (hereinafter — Dolynske Mb.),

which comprises the upper, southernmost part of the Porat Fm.

(Fig. 1). The Dolynske Mb. roughly corresponds to the “Upper

Porat” as distinguished by previous authors. The justification

for the choice of these units will be made in the following

sections of the paper.

Tectonic setting

From the traditional point of view regarding the ancient

deep-seated tectonic elements, the Porat area is located on

the Paleozoic Scythian Platform and a small marginal part of

the Precambrian East European Platform (Fig. 1). To the south

and west, it borders the Jurassic–Triassic North Dobrogea

Orogen and the Mesozoic–Cenozoic SE Carpathians Orogen

with its foredeep. However, it has long been noted that

the Miocene tectonic plan in many cases does not match

the ancient basement structures and the so-called late Miocene

Prutian Trough (at the place of the Porat area) extends east-

ward from the foredeep not less than 150 km (Rudkevich

1955). Based on findings of numerous tectonic deformations

in the terrestrial late Miocene–Pliocene deposits and geomor-

phological analysis, Bilinkis (1992) suggested that the East

European and Scythian platforms in Moldova were re-acti-

vated during the Pliocene as a result of the orogenesis in

the Carpathians.

During the last decades the post-collisional history of

the southern part of the ECF was analysed in terms of

the pro-foreland basin, which appeared as a result of isostatic

compensation and the resultant flexure in front of the former

collisional zone and were filled due to erosional unloading of

the orogen (e.g., Sanders et al. 1999; Leever et al. 2006;

Matenco et al. 2007). The development of the foreland basin is

expressed by asymmetrical subsidence in the late Miocene–

Pliocene, which was replaced by uplift and inversion of

the basin in the Quaternary. Fielitz & Seghedi (2005) as well

as Matenco et al. (2007) extended the foreland basin as far

eastwards as the Prut. The results of our previous (Matoshko

et al. 2016) and present study indicate the existence of

the foreland basin in scale designated by Rudkevich (1955).

The part of the presently reviewed area is located in the foreland

THE PLIO–PLEISTOCENE FLUVIAL SYSTEM OF THE PALEO-DANUBE

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

deposits (e.g., the Kerch iron-ore field; Kholodov et al. 2014),

or the accumulation of abundant glauconite in transgressive

lag deposits (Jorissen et al. 2018). Humidity progressively

increased during the Cimmerian and reached a maximum in

the Kuialnykian, after which the climate became cooler

(Muratov 1964).

The last summary of the landscape–climate features of

the south-western part of the East-European Plain was made

by Svetlitskaya (1994) based on various paleontological

lite rature data, which included plant macrofossils, pollen,

large mammals, small mammals, ostracods, and molluscs.

The climate and vegetation of the regional stages of the latest

Miocene and Pliocene were characterized as follows.

In the Pontian, the area was covered by a forest–steppe

vegetation and the climate was inferred to have been some-

what warmer than now. In the Cimmerian, the area had preva-

lent broad-leaved forests, which became less widespread at

the end of the stage. The early Cimmerian is regarded as

the warmest and probably most humid time in the Pliocene,

whereas the remainder of the Cimmerian is regarded as fairly

dry. In the Kuialnykian, there were mixed conifer and broad-

leaved forests, which were replaced by moist steppe vegeta-

tion towards the east. Climate was initially rather warm and

humid, but a substantial reduction of the forest at the end of

the stage indicates cooling and probably aridification. The trend

of cooling and moderate aridification continued during the Early

Pleistocene (Calabrian), forming the modern steppe landscape

of the area.

Despite some general climatic estimates, the characteristics

of the individual stages of the Pliocene and Gelasian do not

coincide in these studies, which may be due to a different

interpretation of the mixed composition of fauna and flora,

which belonged to different climatic zones.

Methods

We have conducted an intensive review of previous work,

including: the history of the Porat Fm. studies and an essential

refinement of the Pliocene–Quaternary stratigraphy. This was

followed by our own analyses of the lithofacies and sedimen-

tary architecture of the fluvial formations supplemented by

a summary of publications on their mineral and petrographical

composition.

For the purpose of mapping and lithofacies analysis we used

70 outcrops (Fig. 1). Many of them are new, while some were

known from the literature. Most outcrops are several tens of

metres wide and from several up to 30 m high. Thirty of

the largest reference outcrops were subjected to thorough

facies analysis. The position and altitude of the outcrops were

measured using GPS with rough control by topographical

maps and satellite images.

Along with these exposures, we used several dozen bore-

hole descriptions from the state archives (DNVP “Geoinform

of Ukraine” and State Agency for Geology of the Republic of

Moldova). These borehole records were re-interpreted for

an analysis of the sedimentary architecture on formation scale

with construction of cross-sections.

Results

Review of stratigraphy and age of the Porat Fm. and

Dolynske Mb.

Under- and overlying marker beds

In most of the study area, the Porat Fm. erosively overlies

mollusc bearing shallow marine sediments that were deposited

during the early Pontian (Fig. 2). During our study we regis-

tered their direct erosive contact in the Slobodzia Oancea site.

They are thus younger than early Pontian. Magnetostrati-

graphic results from the Dacian Basin and the northern coast

of the Black Sea indicate that the base of the early Pontian falls

in chron C3An.1n and is consequently 6.1 Ma old, while

the top of the early Pontian is 5.8 Ma old (Vasiliev et al. 2004;

Krijgsman et al. 2010).

For deposits more recent than the Pontian, there is a diffe-

rence in stratigraphic terminology between the Dacian Basin

and the Black Sea Basin. In the Dacian Basin, the Pontian is

overlain by the Dacian (Fig. 2). The Pontian–Dacian boundary

is dated at 4.8 Ma (Jorissen et al. 2018). Along the northern

coast of the Black Sea, the Pontian is overlain by the Cim-

merian. The lowermost Cimmerian forms a reddish condensed

interval at the Zheleznyi Rog section on Taman and has a nor-

mal polarity (Vasiliev et al. 2004; Krijgsman et al. 2010). It was

thus interpreted to have accumulated during the Tvera normal

chron with a base at 5.235 Ma (Gradstein et al. 2012). The top

of the Pontian could thus be 5.235 Ma or 4.8 Ma old, depen-

ding on whether a Black Sea or Dacian Basin stratigraphic

framework is chosen.

In the narrow strip south of the New Trotus Fault (Fig. 1)

the Porat Fm. overlies shallow, marine-mollusc-bearing early

Cimmerian deposits (Mokriak et al. 2008). This indicates that

at least part of the Porat Fm. and the complete Dolynske Mb.

are younger than early Cimmerian. Along the lowest Danube,

the Porat Fm. is underlain by fluvial deposits of the Near

Danube Fm. (Matoshko et al. 2009), which was correlated to

the Dacian Stage of the Dacian Basin (Gozhik & Chirca 1973)

and to the Cimmerian Stage of the Black Sea (Mokriak et al.

2008). In this area, the deposits of the Porat Fm. and Dolynske

Mb. are furthermore overlain by estuarine deposits of

the Paleo-Euxinian age (Fig. 2). This sets the upper age limit

for the fluvial formation.

In many places, a weathering crust had developed on

the Pontian before the deposition of the Porat Fm. The top of

the Pontian is moreover frequently coloured in bright red.

At the contact with the overlying alluvial deposits it is accom-

panied by red-coloured conglomerate and red-brown lumpy

frac tured silts-clays interpreted as paleosols (e.g., Ghenea

1968; Roşca 1969; Hubca 1982; Vangengeim et al. 1995;

Gozhik 2006). The Porat Fm. is frequently overlain by clays

MATOSHKO, MATOSHKO and DE LEEUW

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

and silts of red, brown, green, grey and blue tints combined in

present paper in name “Red-Brown Fm.”. This formation

occurs widespread on the Scythian Platform and the southern

part of the East-European Platform. According to Dodonov et

al. (2005) its age range covers the Pliocene and Eopleistocene

(Lower Pleistocene in the current scheme). The origin of

the Red-Brown Fm. is not clear.

Biostratigraphy

Abundant fossil finds from the Porat alluvial deposits are

reported in the local literature. These mainly comprise terres-

trial vertebrates and molluscs. However, a thorough analysis

of the available information on these fossil finds (without

the scrutiny of taxonomy and its knotty history) revealed that

for a significant portion the exact place where the fossils were

found is not documented, or the fossils were not reinterpreted

and classified properly to the modern state of the art. Different

researchers commonly worked on different sections and

obtained different lists of distinguished species that are impos-

sible to match. In our further consideration we have only taken

into account fossils for which the location and classification

were properly documented.

It appears that quite a variety of taxons of large mammals

with overlapping lineages occur in what could be seen as one

vast Pliocene–Calabrian zone (Supplementary Table S1). Others

occupy narrower chronological intervals inside the noted

zone. A more meticulous look, linking the mammal occur-

rences with site locations, highlights rejuvenation of mammal

ages southwards. The youngest species are found at sites

which pertain to the Dolynske Mb. (Fig. 1).

Analysis of the much more numerous remains of small ver-

tebrates and especially the Cricetidae family (Konstantinova

1967; Ali-Zade et al. 1972; Gromov & Polyakov 1977;

Shushpanov 1977, 1980; Alexan drova 1989; Vangengeim et

al. 1995; Radulescu & Samson 2001; Tesakov 2003, 2004)

pro vides more thorough stratigraphic insight confirming

the con clusion made concerning large mammals (Supple-

mentary Table S2). There is an evident up-section younging

trend from European Mammal Zones MN 14 to MN 15 for

the sites of the northern and central part of the Porat area.

There is furthermore a southward younging trend in the fossil

assemblages pertaining to the MN 16–17 zones, including

those found in the Dolynske Mb. It is mainly on the basis of

these fauna variations the previous authors have distinguished

the Lower and Upper Porat units.

More than one hundred species of freshwater and euryhaline

clams are known from the Porat Fm. (e.g., Konstantinova

1967; Ghenea 1968; Sinegub 1969; Ali-Zade et al. 1972;

Gozhik & Chirca 1973; Hubca 1982; Nikiforova et al. 1986;

Gozhik 2006; Andreescu et al. 2013) but only some of them

have a distinct stratigraphic significance. The lowermost beds

of the Porat Fm. especially those directly overlying the Pontian

contain various forms of the Prosodacna genus very peculiar

to the Meotian–Pontian strata of the region. Above these beds,

they are not encountered. The lowest part of the Porat Fm. and

Tuluceşti Fm. are characterized by gauffering Margaritifera of

the Plicatibaphia genus, transforming to smooth forms upsec-

tion (Nikiforova et al. 1986; Andreescu et al. 2013). These are

joined by species of the Psilunio genus and Viviparus bifarci-

natus Bielz. passing up to the middle beds of the Porat Fm.

As a whole they correlate with the Late Dacian. The species

Psilunio sandbergeri Neum. emerged in the upper part of

the Porat Fm. and also occurs in the Tuluceşti Fm. The upper

boundary of the zone with Viviparus bifarcinatus coincides

with the Gilbert–Gauss Paleomagnetic epochs boundary

(Nikiforova et al. 1986).

Paleomagnetic dating

A summary and revision of the magnetostratigraphic study

(Hubca et al. 1983; Sadchikova et al. 1983; Alexandrova

1989) of five sections on the right bank of the Bolshaya Salcia

valley (Fig. 1), accompanied by small mammal identifications,

led Vangengeim et al. (1995) to conclude that the Carbalia

beds pertain to the late part of the Gilbert Chron. They have in

this case used the MN 15 type mammal fauna as an additional

constraint to correlate the polarity pattern of small and isolated

sections. The lower boundary of the formation is placed either

at the top of, or within the Cochiti subchron (C3n.1n; 4.300–

4.187 Ma; Gradstein et al. 2012). The upper boundary of

the formation is considered to be near the boundary between

the Gilbert and Gauss chrons (3.596 Ma; Gradstein et al.

2012). Given the north–south younging trend in the Porat Fm.

demonstrated by the small mammal fauna, sections to the north

of those investigated by Vangenheim may be older, whereas

those to the south may be younger. In addition, the Bolshaya

Salcia sections lay 20–40 m below the local watershed and

may be overlain by younger alluvial deposits that are currently

unexposed.

Because there are no thick, continuous intervals of the Porat

Fm. exposed, paleomagnetic results cannot be correlated to

the regional timescale without the aid of mammal or mollusc

assemblages. Considering the age of the underlying Pontian

and overlying Paleo-Euxinian deposits and the Porat Fm.’s

mammal fauna, which ranges from MN14 to MN17 in age,

single reversals can be correlated to any part of the magnetic

polarity timescale between the middle part of the Gilbert chron

to the lower part of the Matuyama chron.

Lithology and lithofacies of the Porat Fm.

We have used lithofacies analysis to provide reliable criteria

for the mapping of the Porat Fm. including the Dolynske Mb.

and to interpret the sedimentary environment in which they

accumulated. Our outcrop examination was integrated with

the results of previous studies (Konstantinova 1967; Ali-Zade

et al. 1972; Negadaev-Nikonov et al. 1980; Hubca 1982;

Hubca et al. 1983). The determination of lithology, sedimen-

tary structures, textures, their spatial relations and some other

features is used for identification of lithofacies, which are fur-

ther combined into characteristic facies associations. A facies

THE PLIO–PLEISTOCENE FLUVIAL SYSTEM OF THE PALEO-DANUBE

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

association is described here as a stable pattern of distin-

guishable units (consisting of lithofacies), similarly repeated

in a number of sections and bounded by extensive erosional

surfaces.

In general, we distinguished two main groups of facies that

pertain to alluvial and basin environments respectively. There

are two varieties of alluvial associations in the deposits that

are traditionally attributed to the Porat Fm. (Table 1). One of

them is encountered in the majority of the studied sections

within the Prut–Yalpukh interfluve (Fig. 1) and called “Prut–

Yalpukh alluvial facies association”. The other variety occurs

exclusively in several outcrops in a small strip along the lower

reaches of the Danube and Prut rivers near Dolynske (Fig. 1).

The striking difference in lithofacies and the accompanying

difference in occurrence and sand composition, which will be

dwelt on below, are the main reasons to distinguish the latter

as the “Dolynske alluvial facies association”.

Prut–Yalpukh facies association

The Prut–Yalpukh alluvial facies association is represented

by a principal lower sand-dominated unit (A) and a less abun-

dant upper mud-dominated unit (B). Its thickness varies from

2 to 8 m. This structure repeats cyclically. The surfaces that

separate units of different alluvial cycles are the most conti-

nuous surfaces at outcrop, often continuing through the whole

exposure. Up to 4–5 alluvial cycles are registered in the highest

outcrops (Chioselia Mare, Manta, Colibași 1 and Valeni 2)

during this study (Fig. 3). A set of exposures of at least 50 m

height on the right slope of the valley Bolshaya Salcia dis-

played up to 10 such cycles (Hubca 1982; Vangengeim et al.

1995). In places where the Porat Fm. is thicker, more nume-

rous cycles are to be expected. Hubca (1982) wrote about

a general coarsening upwards of sands in the upper layers of

the Carbalia Beds (upper part of the Porat Fm. in the Prut–Yal-

pukh interfluve), which is consistent with our observations.

The A unit was traditionally identified in the area of consi-

deration as “sands with gravel” (Figs. 3, 4). Their granulo-

metry varies from very fine to coarse sand. Fine sand prevails.

There is no analytical data about sand sorting, but visual esti-

mations indicate a predominance of poor and medium sorting

and the absence of fining or coarsening upwards trends.

The A unit is dissected by numerous subhorizontal, slightly

concave and sometimes incision-like scour surfaces. Among

them we distinguish the main intra-unit type of surfaces which

separated two varieties of the A unit differing by sedimentary

structures and content of coarse material.

The first variety (A1, the most common) is dominated by

cross-bedded (trough and planar) and ripple cross-laminated

sands. The second variety (A2) contains parallel (or semi-

parallel) laminated and trough cross-bedded sands sometimes

with a unidirectional dip of the cross-sets (Figs. 3, 4). The second

variety also differs by more irregular structure with greater

abundance of scours, more frequent soft sediment deforma-

tions and usually an increased amount of gravels, which can

be up to 2 m thick at the unit base.

Sediments of both the A1 and A2 varieties occur in rela-

tively thin (generally 0.2–2 m) and laterally limited (usually

a few meters long) layers or lenses replacing each other in

an irregular way. Their cross-bedding sets are 0.1–0.3 m thick

on average, reaching in places up to 1 m. The dip of foresets in

the A unit shows that paleocurrents varied greatly from west

over south to east, with an average southerly direction.

The A unit contains scattered gravel clasts and clusters thereof

in the coarse sand matrix part of channel lags that line the ero-

sional base of the unit. There is a noticeable increase in gravel

clasts in the northern part of the study area and in sites close to

the Prut River from Manta (upper part of section) southwards

(Fig. 1).

Gravels primarily include carbonate nodules, mud clasts as

well as exotic clasts. Mammal bones and mollusc shells (pre-

dominantly freshwater) are encountered in many exposures,

but clusters of shells (coquina) are found only in the basal

horizons of the Colibași–Brînza and Valeni–Slobodzia Mare 1

sites.

Soft sediment deformation is very common, particularly

convolute bedding. It is frequently associated with clustered

Skolithos-type trace fossils, probably representing escape

traces that could made following soft sediment deformation.

The average full thickness of the A unit varies from 4.5 m in

the north (e.g., Ciobalaccia, Chioselia Mare, Musaitu, Pelinei)

to 3 m in the south (e.g., Colibași 1, Valeni 2, Vulcănești 1–3,

Bolgrad 1). The full thickness of the A unit is considered mea-

surable only in cases where it is overlain by the B unit. Where

anomalously large thicknesses (6–10 m or more) are men-

tioned in published sections, these probably represent amal-

gamated deposits, which include one or two partial alluvial

cycles.

Facies association

Lithofacies

Thickness

Process interpretation

Prut–Yalpukh

unit

Sandy silts to clays, either structureless or with parallel lamination.

They often show mottling and include carbonate nodules.

0–6 m

Overbank deposition alternating with

subaerial exposure.

unit

Sands with gravel showing cross-bedding (trough and planar), ripple

cross-lamination (cross-bedding sets are 0.1–0.3 m thick on average,

reaching in places up to 1 m) and parallel lamination. They are

dissected by numerous scour surfaces and often include soft

sediment deformation.

3–5 m

Migration of 3–5 m deep channel with

moderate to strong variations of discharge.

Dolynske

Sands with bottom gravel, showing cross-bedding, ripple cross-

lamination and parallel lamination.

10–20 m

Migration or incision of 10–20 m deep

channel with persistent discharge regime.

Table 1: Characteristic of facies associations.

MATOSHKO, MATOSHKO and DE LEEUW

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

The B unit lies above unit A with a clear, in places sharp

contact. It varies from sandy silts (B1 lithofacies) to clays (B2

lithofacies), either structureless or with parallel lamination,

usually finning upwards and is up to 4–6 m thick (Fig. 5a).

However, in many cases it is less thick or absent. Its lower

contact with the A unit is either sharp, gradational or rarely

through alternation. Muds are characteri-

zed by greenish-orange mottling and fre-

quent occurrence of patchy or layered

carbonate nodules (prevailing in clayey

variety). Rarely the top of the B unit con-

sists of a paleosol up to 2 m thick (e.g.,

Huluboaia-Doina, Musaitu, Vladimirovca,

Etulia Noua). Hubca (1982) also sug-

gested frequent occurrence of hydromor-

phic soils in sections of the Bolshaia

Salchia River, but indicated only an ele-

vated organic content as proof. In two

cases (Vulcănești 3, Borceag) we obser-

ved buried soils with a full profile under-

lain and overlain by unit A (Fig. 5b). More

often the B unit does not reveal signifi-

cant pedogenic alteration. In this unit

there sometimes is some wood detritus,

but in general plant residues are very rare

in these deposits.

Unit A is usually directly overlain by

facies of unit B, or vice versa, in the Porat

Fm. However, in some rare cases at the top

of the outcrop, unit A facies are overlain

by silts and muds, with thinly interbedded

fine sands. They are clearly separated by

gently inclined-concave surfaces that

repeat laterally (Fig. 5a, c).

The relatively thick erosionally based

sands of unit A, dominated by current

cross-bedding, cross-lamination and often

inclusions of the freshwater mollusc

shells (including rheophilic ones) are

interpreted as channel deposits. The chan-

nel deposits of the A1 variety fit classical

descriptions (e.g., Miall 2006; Bridge &

Demicco 2008) best and imply a rela-

tively even discharge regime with only

Fig. 4. The Prut–Yalpukh facies association,

unit A: a, b, c — fragments of the basal hori-

zon containing typical assemblages of mud

clasts (greenish), carbonate nodule clasts

(whitish) and rock clasts (brown and black

cherts, light grey quarts and sandstones):

a — dominated by mud and carbonate nodule

clasts (Vulcăneşti 2), b — dominated by mud

clasts (Borceag), c — dominated by rock and

carbonate nodule clasts (Manta); d — upward

succession of different lithofacies of unit A:

1 — sands with trough cross-bedding,

2 — sands with planar cross-bedding,

3 — lense of gravelly sands, 4 — sands

with sub-parallel lamination (Chioselia Mare).

The comb is 14 cm and shovel 1 m long.

Fig. 3. Porat Fm. Cycles of regularly stacked alluvial facies associations separated by ero-

sional surfaces and overlain by Red-Brown Fm. at the Valeni 2 site. For the lettering of units

see the text.

100

MATOSHKO, MATOSHKO and DE LEEUW

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

only in a few of the southernmost outcrops (Dolynske 1–4,

Fig. 1). Similar deposits were also described before from

the Giurgiuleşti 2 and Lymanske 1 outcrops (Konstantinova

1967). These alluvial deposits are characterized by a unit

(10–20 m thick in exposures) of alternating (0.5–2 m thick)

beds of trough cross-bedded, ripple cross-laminated and occa-

sionally parallel laminated sands showing great textural uni-

formity and containing a significant amount of mica (Fig. 6).

The cross-bedding sets are generally 0.3–1.0 m thick with

a maximum up to 1.5 m. The cross-bedding dip clearly

indicates a predominantly easterly paleocurrent direction.

Rather common for these strata are sandstone nodules

occurring parallel to the bedding. The cementation suggests

high carbo nate content. According to (Konstantinova 1967)

the basal gravels are occasionally observed and reach 4 m

thickness. We distinguish these deposits as channel facies using

the same principals concerning A unit of the Prut–Yalpukh

facies association. However, unlike the cyclically built Prut–

Yalpukh association, the deposits of the Dolyn ske Mb., dis-

tributed along the lowermost Danube, are uninter rupted.

Basin facies

In some southern outcrops along the lower Prut and along

the lowermost Danube (Valeni–Slobodzia Mare 2, Dolynske

1, 3) the sandy alluvial deposits are overlain by thick planar-

bedded mudstones that are clearly different from floodplain

deposits. At the Dolynske sites, they consist of thick and uni-

form, thinly laminated clays (Konstantinova 1967). Lamination

is either plane parallel or wavy. The transition from the under-

lying sands of the Dolynske alluvial facies association occurs

through micaceous silts with the same sedimentary structures

and nodules that are characteristic for the sands. The total

thickness of the exposed muds reaches 19 m. At the site Valeni

–Slobodzia Mare 2, very similar muds overlie alluvial depo-

sits of the Porat Fm. (Fig. 7). The transitional interval is

marked by micaceous fine sands with nearly unidirectional

ripple cross lamination indicative of northerly paleo-flow,

possibly deposited in an estuarine environment. The basal

mudstones contain some minor burrows and some sporadic

freshwater ostracods (pers. comm. M. Stoica). We interpret

these muds to have been deposited in a calm lacustrine envi-

ronment below the wave base.

According to Rengarten & Konstantinova (1965), characte-

ristic drapes of coal detritus and mica particles oriented in one

direction line the horizontal lamination of well sorted sands,

silts and clays. They also noted that some of the bluish-grey

uniform structureless clays contain rare foraminifers and

grains of authigenic glauconite (0.05–0.1 mm). Rengarten &

Konstantinova (1965) infer from these observations that

the muds accumulated in a brackish basinal environment.

Drilling in the southern part of the research area revealed

that there is a thick interval with muds (Gozhik & Chirca

1973; Gozhik 2006), similar to those exposed at outcrop near

Dolynske, Valeni and Slobodzia Mare. However, in line with

our own observations and contrary to those of Rengarten and

Konstantinova, the molluscs from these deep horizons indi-

cate that they have accumulated in a freshwater environment.

Mineral and petrographic composition of the Porat Fm.

Sands

According to Hubca (1982) the sands of the Carbalia Beds

(here seen as the lower and middle parts of the Porat Fm.) are

quartz-rich (70–75 %), but with a noticeable content of feld-

spars (up to 10–11 %) and fragments of cherty rocks (10–16 %).

The key assemblage of heavy minerals of the lower part of

the Porat Fm. (Carbalia, Musaitu, Tătăreşti, Cahul and Etulia

Fig. 6. The Dolynske facies association. Alluvial deposits characte-

rized by monotonous metre-scale interfingering of horizontal lami-

nated and cross-bedded mica-rich sand: a — with predominance of

cross-bedding (Dolynske 2); b — with predominance of horizontal

lamination with numerous sandstone nodules (Dolynske 4). The shovel

is 1 m long.

101

THE PLIO–PLEISTOCENE FLUVIAL SYSTEM OF THE PALEO-DANUBE

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

Nouă) in the fraction 0.25–0.01 mm is very stable: garnet

(30–50 %), ilmenite (10–26 %) and leucoxene (10–25 %),

with a total percentage more than 50 %. The heavy fraction

(0.1–0.01 mm) of the Porat Fm. at Cîşlița–Prut also consists of

50–60 % of opaque minerals and garnet (Rengarten &

Konstantinova 1965). In BăIăbăneşti, opaque minerals com-

prise up to 75 % of the heavy mineral fraction, while garnet

takes the second place (Ghenea 1968). In some localities near

the base of the Porat Fm. at the contact with the Pontian,

an increased content of hornblende, andesine, volcanic glass

and some other minerals occurs. This has been considered to

be due to volcanic activity (Hubca 1982).

Apart from the presence of a significant proportion of

opaque minerals, the here observed garnet abundance is com-

mon in the EC foreland deposits. For example, it was noted in

the late Miocene Balta Fm. (Matoshko et al. 2016) as well

as for the sand of the present-day Prut and Dniester Rivers

(de Leeuw et al. 2018, Supplementary material).

The proportion of heavy minerals in the fraction 0.1–0.01 mm

of the sands of the Dolynske Mb. varies from fractions of

a percent (probably lacustrine deposits) up to 16–18 % (pro-

bably channel lag deposits) (Rengarten & Konstantinova 1965).

There is a striking predominance of hornblende and epidote

(26–48 % both), alongside opaque minerals (27–34 %)

(Rengarten & Konstantinova 1965). In addition, sand of

the Dolynske Mb. (Giurgiuleşti 2, Dolynske and Limanske

sites, Fig. 1) is rather micaceous, including dark coloured mica

of the biotite group and colourless mica. This has previously

been reported by Rengarten & Konstantinova (1965) and

Hubca (1982), and was confirmed for the Dolynske and

the lower part of the other outcrops in the course of the present

study. The detrital fraction of the sand is polymict and includes

quartz, feldspar, fragments of flints and schists.

Gravels

Gravel is a permanent intrinsic component of the Porat Fm.

sandy facies. The gravels contain local and reworked far-trans-

ported material. The first group consists of rounded clay balls,

clayey–silty and carbonated nodules. The second group is rep-

resented by relatively hard rocks such as limestones, rare

sandstones and the so-called “Carpathian Pebbles”. In most

cases the mud clasts prevail.

The “Carpathian Pebbles” are a distinctive far-transported

feature for the whole Porat Fm. These “pebbles” are small

slightly rounded fragments and grains (1–2 mm to 3–5 cm

intercept) of brown-red, yellowish jaspers, flints and silicified

claystones. The features of these pebbles have been elucidated

in detail by Matoshko et al. (2016). Their primary sources

have not been accurately established. A secondary source for

these pebbles in the Porat Fm. is the late Miocene Balta Fm. to

the north, part of which was being eroded during the Pliocene

and Pleistocene.

Sands with a high content of gravel represented by meni-

lites, sandstones and some other fragments occur in the BăIă-

băneşti and Tuluceşti sites (Ghenea 1968). Our observations

near Bereşti and Măluşteni have shown that the menilites

reported in the Romanian literature and the “Carpathian

Pebbles” reported in the Ukrainian literature are the same type

of rock. According to Ghenea (1968), at some outcrops of

the BăIăbăneşti Fm. (which we consider part of the Porat Fm.

here), the percentage of menilites is 35–37 %, quartz 27–30 %,

sandstone 29–30 %, and quartzite 4 % in the fraction of gravel

larger than 1 cm.

Gravelly channel lags of the Dolynske Mb. were only infre-

quently observed during the course of this study, because

the lower parts of the corresponding outcrops are at present

poorly exposed. These lags include the “Carpathian Pebbles”.

Rengarten & Konstantinova (1965) mention that there are in

addition fragments of granites, metamorphic schists, and

quartzites associated with the micaceous sands. These authors

attribute this to a direct influence of the adjacent North

Dobrogea Orogen. Except for mud balls (Fig. 4a), most other

clasts do not exceed 2–3 cm across and large pebbles were

encountered in two locations.

Muds

While the muds of the Porat Fm. are often called clays in

the literature (e.g., Rengarten & Konstantinova 1965), detailed

descriptions as well as our own field data reveal that most of

them are silts with variable admixtures of clay or sand. Real

clays of the Porat Fm. consist of 91.3 % of fraction less than

0.01 mm, 8.3 % (fraction: 0.1–0.01 mm) and 0.4 % of fraction

more than 0.1 mm (Zheru 1978). The Porat clays comprise

a hydromica–montmorillonite assemblage with a noticeable

admixture of mixed-layer minerals (Hubca 1982; Mokriak et

al. 2008). The authigenic components in the muds are carbo-

nates, iron and manganese. An increased content of calcite and

dolomite in nodules and in dispersed form (up to 5 %) is

attributed to the lacustrine origin of the muds by Hubca (1982).

In some places, carbonates furthermore cement sandy and

Fig. 7. The basin (probably lacustrine) facies of the Porat Fm.,

horizontally laminated uniform silts with unidirectional ripples at

the base, lying with sharp contact on the A unit of the Prut–Yalpukh

facies association (Valeni–Slobodzia Mare 2). For the lettering of

units see the text.

102

MATOSHKO, MATOSHKO and DE LEEUW

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

gravelly channel deposits. A somewhat different description of

the lacustrine deposits near Dolynske was cited by Rengarten

& Konstantinova (1965) and briefly in the section “Basin

Facies”.

Sedimentary architecture of the Porat Fm.

The Porat Fm. within the SE Carpathian foreland

Abundant archive geological survey data from the Ukrainian

and Moldavian territories, in particular geological maps and

borehole descriptions, were used to supplement the informa-

tion obtained from outcrops. This provides a more complete

picture of the general sedimentary architecture of the Porat

Fm., which we synthesize with the stratigraphic, lithofacial

and lithological data provided above, as well as some geomor-

phological observations.

According to geological maps (Ghenea & Ghenea 1967;

Mokriak et al. 2008), as well as our own data, the Porat Fm.

has a triangular shape in plan, with its top at Cociulia, a site

located near the supposed northern boundary of the Pontian

shallow marine deposits (Fig. 1). The thickness of the Porat Fm.

within the studied Bârlad–Prut–Yalpukh interfluves varies

from 40 to 70 m (except the Reni–Izmail trough, see below);

its average basal slope in the south-south-east direction is

about 0.0028 (0.160). The formation wedges out to the north

and east.

Pontian shallow marine deposits underlie the Porat Fm.

almost throughout the area and are inclined slightly to the S

and SSW. The Pontian–Porat contact also dips in this direction

(Fig. 8). The contact has the character of a terrestrial break in

deposition as indicated by a weathering layer at the top of

the Pontian rocks in several outcrops (Hubca 1982; Van gen-

geim et al. 1995). In other places there is no obvious evidence

of protracted exposure before deposition of the Porat Fm.

In many places the Porat Fm. is covered by the Red-Brown

and Loess formations. Within the Prut–Yalpukh interfluve

these are up to 58 m thick, decreasing to the east and towards

the modern valleys. Along its south-western boundary,

the Porat Fm. disappears into the subsurface and is overlain by

younger Pleistocene–Holocene strata.

Along the triangle’s north-west corner, somewhere in

the Siret–Bǎrlad interfluve, the Porat Fm. is supposedly late-

rally replaced by Dacian age littoral to delta-front deposits

(Jorissen et al. 2018), Romanian age fluvial deposits (van

Baak et al. 2015) and Early Pleistocene (Calabrian) alluvial

fan conglomerates of the Cȃndeşti Fm. (Andreescu et al.

2013). This western lateral contact was not observed in

the field, but should exist considering the stratigraphic posi-

tion of the Porat Fm.

A number of small-scale rootless deformations (folds and

faults with amplitude up to 20–30 m) of non-sedimentary

nature were distinguished within the Prut–Yalpukh interfluve

in the Pontian rocks and the Porat Fm. in the walls of gullies

and in borehole sections by Bilinkis (1992). This observation

was confirmed during the present study. There are also several

examples of disjunctive deformations in the Pliocene–

Quaternary strata of the Bârlad–Prut interfluve (Matenco et al.

2007).

The Porat Fm. in the Reni–Izmail Trough

For the current study, the most interesting boundary of

the Porat Fm. is situated in the southeast, where the formation

enters the late Miocene–Pliocene Reni–Izmail Trough (intro-

duced by present authors). This trough is located between two

main faults: the New Trotus and St. George faults (Fig. 1). It is

formed by a set of successive normal faults, which were active

during and after the deposition of the Porat Fm. (Fig. 8).

The trough evidently tapers downwards on the Dobrogea side.

Its southern flank is represented by epimetamorphosed

Precambrian strata and sedimentary formations of Paleozoic

and Mesozoic age, while the northern flank is formed by late

Miocene (Sarmatian, Meotian and Pontian) deposits. The base

of the formation, as well as the underlying brackish-marine

formations clearly step down across these faults.

To the north of the New Trotus Fault, the Porat Fm. overlies

Pontian shallow-marine deposits (Fig. 8). South of the New

Trotus Fault, and so inside the trough, the Porat Fm. overlies

Cimmerian (Early Pliocene) shallow marine sediments

(Mokriak et al. 2008). The Porat Fm. here includes thicker

mud intervals. In the deepest part of the trough the Porat Fm.

(80 m thick) lies on the Cimmerian age fluvial Near Danube

Fm. (Matoshko et al. 2009). The base of the Porat Fm. is there

located lower than observed further north, outside of the trough,

but its position is shown at different altitudes in borehole

descriptions and cross-sections (Gozhik & Chirca 1973;

Gozhik 2006; Matoshko et al. 2009).

The Near Danube and Porat formations both consist of sands

and gravels and interfinger with muds containing brackish and

freshwater mollusc fauna (Gozhik & Chirca 1973; Mokriak et

al. 2008). Towards the Black Sea, within the Danube Delta,

some boreholes found Levantine fauna at a depth of 68–100 m

(Chirca 1969, with reference to Litianu et al. 1963). This could

indicate that the observed trough and Porat Fm. continue into

that area.

The Dolynske Mb. in the Reni–Izmail Trough

The Reni–Izmail Trough is also noteworthy for the appea-

rance of the Dolynske Mb. The Dolynske Mb. is up to 40 m

thick, occurring alongside the adjacent Prut–Yalpukh type

deposits of the Porat Fm. and overlying Cimmerian age clays

(Figs. 8, 9). Archive data as well as literature (Gozhik 2006;

Matoshko et al. 2009) show that its base lies from 5 m above

sea level to 40 m below sea level on the differently lowered

blocks of the Reni–Izmail Trough. At its northern margin it is

located somewhat lower than the base of the directly adjacent

Porat deposits (Fig. 9). In the axial part of the Reni–Izmail

Trough at the south-western tip of the Yalpukh Lake,

the Dolynske Mb. directly overlies older deposits of the Porat

Fm. The contact is located at 40 m below sea level and lined

103

THE PLIO–PLEISTOCENE FLUVIAL SYSTEM OF THE PALEO-DANUBE

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

Fig. 8.

Geological

cross-section

Cociullia–Isaccea

based

materials

DNVP

“Geoinform

Ukraine”

and

State

Agency

for

Geology

the

Republic

Moldova

(with

small

reinterpretation),

Sinegub

(1969),

Bilinkis

al.

(1976),

and

data

present

study

. Global

units:

—

Holocene;

—

Pleistocene,

P2–3

—

Middle–Upper

Pleistoce

ne;

—

Middle

Pleistocene,

—

Lower

Pleistocene,

—

Pliocene,

—

Miocene,

Pz–Kz

—

Paleozoic–Cenozoic,

—

Paleozoic.

Local

units:;

—

Paleo-Euxinian

Fm.,

—

Red-Brown

Fm.,

—

Porat

Fm.,

—

Dolynske

—

Near

-Danube

Fm.,

—

Cimm

erian,

—

Balta

Fm.,

—

Cahul

Fm.,

—

lower

Pontian,

p1-km

—

lower

Pontian–Cimmerian

undif

ferentiated,

—

alluvial,

l-a

—

lacustrine-alluvial,

est — estuarine, e-sw — eolian-slopewash. Location of the cross-section see in Fig. 1.

104

MATOSHKO, MATOSHKO and DE LEEUW

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

by a 2–4 m thick layer of gravel. The deposits of the Dolynske

Mb. are traced by single borehole sections to Izmail and from

Izmail 70 km further eastwards (mapped by Cherednichenko

et al. 1985) up to the spit of the Sasyk Liman (Palatnaia 1991;

Gozhik 2006) and very probably grade laterally into similarly

aged deposits of the Dniester valley.

The deposits of the Dolynske Mb. are overlain by basin

muds (Fig. 9). These basinal deposits are 8–10 m thick and

have their base at roughly 30 m above sea level. The basinal

deposits on top of the Porat Fm. are in turn overlain by clays

and silts of the Red-Brown and Loess formations while the basi-

nal deposits above the Dolynske Mb. are almost only covered

by a loess blanket.

The geological survey maps (archive data) demonstrate that

the youngest Pliocene (late Gelasian–Calabrian in our inter-

pretation) alluvial deposits (probably an age equivalent to

the Dolynske Mb.), occur on both banks of the modern Cahul

and Yalpukh estuaries, with a marked exposure near Vino-

gradivka. In this connection it is very likely that Porat deposits

with Prut–Yalpukh facies association of a similar age to

the Dolynske Mb. stretch along the banks of

the Lower Prut along a narrow strip (Fig. 1).

The alluvial deposits in outcrops Colibași–

Brînza, Valeni and further south to Giurgiuleşti

are not only associated with younger fauna and

the presence of coarser material (see above),

but it also occurs in the same visual narrow alti-

tude interval as exposures near Dolynske (up to

25–30 m), being overlain only by loesses or by

thin clays of the Red-Brown Fm. (Valeni 2).

Pliocene–Pleistocene river terraces

Konstantinova (1967) suggested a staircase-

type stratigraphy in the valley of the Prut River,

with ten terraces deposited by the ancient Prut

River. The Upper Porat Unit, bearing the youn-

gest mammal fossils, was interpreted as the upper-

most river terrace, inset into the Lower Porat unit

represented by a thick package of aggradational

alluvial deposits. The distinction of the terraces

was based on fauna (see above), but not on any

analysis of the geomorphology. Nevertheless,

this concept was apprehended by some resear-

chers (Sinegub 1969; Bukatchuk et al. 1983),

whereas Chirca (1969) concluded that there were

no reliable reasons for the identification of at

least the 8

, 7

and 6

terraces.

According to our own paleo-geomorphologi-

cal interpretation, the top of the Porat Fm. formed

a vast alluvial plain into which the Dolynske

Valley was incised. The Porat Fm. and its

Dolynske Mb. were then buried under the Red-

Brown and Loess formations. At present all of

them form a common lowland surface that gra-

dually descends in a southward direction. Along

the lowest reaches of the valleys of the Danube,

Siret and Prut, there are one or two Middle–Late

Pleistocene and Holocene well-expressed river

and estuarine terraces above the floodplain at

6–15 m above sea level as well as a marine ter-

race with a rich Old Euxinian, Middle Pleistocene

mollusc fauna at 25–35 m above sea level (Saulea

et al. 1967; Negadaev-Nikonov et al. 1980;

Bukatchuk et al. 1983). Above the Old Euxinian

terrace, there are no levels resembling terraces in

Fig. 9. Geological cross-sections through the alluvial Dolynske Mb. and its correla-

tion with the rest of the Porat Fm. near Reni, based on materials of the DNVP

“Geoinform of Ukraine” and observed exposures. Sedimentary environments are

according to our interpretation. DA — Dolynske alluvial facies association, charac-

terized at exposures. See position of the inset map in Fig. 1.

105

THE PLIO–PLEISTOCENE FLUVIAL SYSTEM OF THE PALEO-DANUBE

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

the modern surface near the main rivers. This can be explained

particularly by intensive post-Porat erosion and plunging of

the Pleistocene alluvial deposits below sea level, which was

revealed in other valleys of the Black Sea north-western coast

(Matoshko et al. 2009).

Discussion

Sedimentary environment

The rivers of the Porat Fm.

There are several features of the Prut–Yalpukh alluvial facies

association (see above) which we consider key for our recon-

struction of the rivers that deposited the Porat Fm. These are:

• the sand with gravel composition of the channel facies;

• the fractional lenticular-layered irregular structure of the

channel units;

• the abundance of intra-unit incisions in channel deposits;

• the absence of vertical trends in granulometry of the channel

deposits;

• the relatively small thickness of the cross-bedding series;

• the sharp contact between channel and floodplain facies;

• the simple vertical structure of the floodplain unit, including

one-two varieties;

• the almost complete lack of abandoned channel facies.

In combination, these features indicate a characteristic

sandy high-energy braided river (type “D” in the classification

of Rosgen 1994). It is supposed that the braided system of

channels (braided belt) was active during fluvial seasons or

flood episodes accompanied by frequent shifts of the river

branches. The maximum bankfull channel depth could be esti-

mated based on the thickness of the channel deposits (Bridge

2003). This suggests on average 3–5 m deep channels. Their

width was, according to width/depth ratio > 40 for channels of

type “D” (Rosgen 1994), about one to two hundred metres and

they had a medium slope. The shape of the channels resembled

a shallow trough with gentle flanks. The abundance of dewa-

tering structures suggests high rates of deposition.

The alluvial cycle consisted of channel erosion, fast channel

infill followed by rapid covering by floodplain deposits.

Flooding of the river plain alternated with subaerial exposure

as indicated by soils, reddish muds, carbonate nodules, etc.

General features of the Porat Fm. sedimentation

We interpret the deposits of the Porat Fm. to have been

deposited by a wide braided belt with numerous intertwined

equal branches. This is confirmed by the observed facies asso-

ciations, which are indicative of very stable channel condi-

tions throughout the area and through time, as well as

the pre servation of downstream discharge. With the lapse of

time, the braided belt wandered radially over the coastal plain,

around the main south-south-west axial direction. Within

the study area, we did not see any replacement of alluvial

deposits by deltaic facies, which means that sea-ward pro-

gradation of the fluvial system was strongly sediment

supply-driven.

The vertically stacked alluvial cycles, which repeat through-

out the whole Porat Fm., provide evidence of continuous (up

to 2 Ma) repetition of the environment of the elementary cycle

described above. The steady aggradational trend caused reduc-

tion or full scour of the floodplain units as well as part of

the channel units, which is common for the aggradational

fluvial systems of the East European Plain (Matoshko et al.

2004). This also indicates that sediment supply was equal to,

or larger than the local rate of creation of accommodation

space. Sand and gravel compositions indicate that the Car-

pathians and the Balta Fm. provided much of the sediment that

accumulated in the Porat river basin. There was also some

influence from local sources: the Pontian–Cimmerian strata

underneath the Porat Fm. The role of this source reduced

through time due to aggradation of the alluvial deposits.

The wandering and seaward progradation of the Porat flu-

vial system as well as aggradation of the alluvial strata resulted

in the particular geometry of the sedimentary body of the Porat

Fm., which resembles a flattened cone with the top at the mar-

gin of the sedimentary Dacian Basin. This shape satisfies one

of the main features of “alluvial fan” in definition of Miall

(2000). However, according to our facies interpretation,

the Porat River did not diverge into radiating distributaries

along its course (second feature in this definition). It therefore

cannot be strictly referred to the so-called “fluvial distributary

system”. Without delving into the unsettled issue of termino-

logy associated with fan-shaped alluvial bodies (e.g., Miall

2000; Nichols & Fisher 2007), we decided to use a neutral

term and designate the Porat Fm. as an alluvial basin infill.

The river of the Dolynske Mb.

A rather different sedimentary environment was responsible

for the deposition of the Dolynske Mb. The large thickness of

its single river channel deposit (10–20 m), with cross-bedding

sets of up to 1.5 m, is indicative of a really large and powerful

river (much more powerful than the distributary streams of

the Prut–Yalpukh type) with a persistent character of high dis-

charge and a very high input of bedload. This river could be

referred to “entrenched (gully), step/pool and low width/depth

ratio on moderate gradient rivers of the type G” ( Rosgen 1994).

Near Dolynske it cut a valley about 20 km wide and filled it

with a thick channel deposit, which we call the “Dolynske

valley infill”. In contrast with the rest of the Porat Fm. it was

a one-time, rather than cyclical process.

The Dolynske Mb., moreover, has a very different sand

composition from the rest of the Porat Fm. and we accordingly

infer a different provenance. The heavy mineral assemblage of

the Dolynske Mb. (hornblende–epidote, opaque minerals) is

similar to that of the present-day Danube and to those of rivers

in the western Dacian Basin, which are also rich in epidote and

have a comparatively small garnet component (de Leeuw et al.

2018). Epidote is, on the contrary, very scarce in the sediments

106

MATOSHKO, MATOSHKO and DE LEEUW

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

of the Balta and Porat formations that were sourced from

the Outer Carpathians, while these sediments contain a very

predominant garnet component. We interpret the Dolynske

Mb. as a reflection of the appearance of the Danube along

the southern margin of our study area at the beginning of

the Late Porat stage.

Probable basin environment

The scant information provided here about the basin facies

still suggests the periodic appearance of a lacustrine environ-

ment in the near-Danube area and especially in the Reni–Izmail

Trough prior, during and after accumulation of the Porat Fm.

(Fig. 2). The muds overlying the alluvial deposits of the Porat

Fm. and Dolynske Mb. indicate the final episodes of basin

deposition in that area, in combination with a steady input of

suspended load. The basin was a predominantly freshwater

environment with possible phases of salinization. This is not

surprising considering that the Black Sea was a lake during

most of the Pleistocene (Ross 1978; Krijgsman et al. 2019).

It is important to underline that the fluvial facies of the Porat

Fm. are abruptly replaced by basinal mudstones, without

deposition of intervening deltaic facies.

Comparison with other fluvial series along the northern

Black Sea coast

The Porat Fm., including its Dolynske Mb., is not an excep-

tional phenomenon along the northern coastal plains of

the Black Sea and Sea of Azov. Earlier Konstantinova (1967),

Tchepalyga (1967), Bukatchuk et al. (1983), and Matoshko et

al. (2009) correlated it with other fluvial formations. These

include the Lower Dniester terraces (Fîrlădeni, Hagimus and

Kitskany, or some of them) and the ancient alluvial deposits

near the Dniester Liman, as well as the Kuchurgan Beds and

Kuialnykian Series. All of these were deposited in the same

chronological interval. Most of them are associated with

powerful rivers, the precursors of the Dniester, Southern Buh

and Dnieper. These fluvial formations embrace vast areas,

have a comparable thickness and aggradational structure in

their distal seaside part, as well as some of them also having

a fan shape. Thus, their origin is due to a common regional

cause or causes such as tectonics, evolution of the associated

basins, landscape and climate.

Tectonics as an important factor in fluvial development

From a tectonic point of view, the subsidence in the Dacian

Basin, with its main depocenter in the Focsani Depression

(Necea et al. 2005; Leever et al. 2006; Matenco et al. 2007;

Jipa & Olariu 2009), generated the accommodation space for

the Porat Fm. During the late Zanclean, Piacensian and early

Gelasian, the subsidence rate of the basin was in close balance

with a weak uplift of the river drainage area to the north of

the basin (i.e. in the ECF). The fringe zone between the basin

and the uplifting area gradually shifted to the S and SSW.

The local rate of subsidence was in balance with or smaller

than sediment supply from the Outer Carpathians and the ter-

restrial plains upstream of the Porat area, leading to aggra-

dation of the alluvial cycles. The lack of deltaic facies

fun da mentally distinguishes the Porat Fm. from the progra-

ding system of the Balta Fm., which is characterized by a full

set of facies — from typically alluvial through deltaic to basin

ones (Matoshko et al. 2016). During the Porat Stage of depo-

sition, deltas were likely located just beyond the southern

margin of our study area, namely in the subsurface along

the retreated northern margin of the Dacian Basin.

The incision that cut the valley in which the Dolynske Mb.

was deposited could have been stimulated by a slowdown of

the subsidence, but also by changes in water level in the basin

(see below). Subsidence resumed at its previous rate during

the Calabrian which coincided with the deposition of

the Dolynske Mb. Finally, the post-Dolynske incision (second

half of the Calabrian) may have been aided by cessation of

subsidence in our study area, and the onset of the general uplift

and tilting of the former basin area in the ECF with appearance

of highlands and deep incised valleys instead of former flat

lowland. This tectonic evolution resembles that of the NW

margin of the Focşani Basin, which subsided during the Plio-

cene, was uplifted during the late Pliocene, resumed its sub-

sidence during the earliest Pleistocene, and experienced a final

period of uplift in the late Early Pleistocene (Necea et al.

2005).

The local tectonic influence for the Porat Fm. is related to

the Reni–Izmail Trough; an active structure with block-type

subsidence. An increase in thickness of the late Miocene and

Pliocene units in the Reni–Izmail Trough across subsequent

faults reveals the syndepositional character of the normal

faulting (Fig. 8). Despite the additional accommodation space

in comparison with northern regions, fluvial incisions did

occur, as manifested by scoured contacts and the presence of

gravelly basal horizons.

The features of the Reni–Izmail Trough and the presence of

local rootless displacements in the Porat Fm. provide evidence

for tectonic activity of the eastern part of the SE Carpathian

foreland, which penetrated deeply into the platform region

during the Pliocene and probably the Calabrian. This means

that the SE Carpathian foreland basin was still in its construc-

tive phase at this time, in line with fission track evidence from

the adjacent part of the orogen (Sanders et al. 1999).

Porat Stage: fluvial system evolution in connection with

basin history

During the late Miocene, the general trend of development

of the ECF was characterized by replacement of the offshore

environment by a terrestrial one. The latter was intimately

associated with the evolution of fluvial systems, which came

into being in basin-margin settings along the periphery of

the Eastern Paratethys. The latter part of the evolution of

the ECF is known as the Balta Fluvial Stage (Matoshko et

al. 2016). The Balta Stage was followed by a marked

107

THE PLIO–PLEISTOCENE FLUVIAL SYSTEM OF THE PALEO-DANUBE

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

transgression in the Early Pontian and subsequent retreat of

the Paratethys from the East Carpathian foreland.

The next Porat Fluvial Stage began in the early Cimmerian.

We distinguish the Near-Danube, Early Porat and Late Porat

substages; they correspond to the Near-Danube and Porat for-

mations as well as Dolynske Mb. (Fig. 2).

Near-Danube Substage: incision and its subsequent filling

On an early map by Muratov (1964), the Dacian and Black

Sea basins were connected by a hypothetical river through

which freshwater of the Dacian Basin flowed into the brackish

Black Sea during the Cimmerian. This river has subsequently

been called the “Galaţi Passage” (Saulea et al. 1967), simply

the “strait” (Mokriak et al. 2008), the “Reni Strait” and

the “Reni Sill” (Popescu et al. 2009), the “Scythian Gateway”

(Munteanu et al. 2012) or the Barlad Strait (Palcu et al. 2017).

While considering this area very important for the connec-

tion–disconnection history of the Dacian and Black Sea basins,

none of these authors actually cited data on its Pliocene–

Quaternary geology. As we shall show below, offshore, terres-

trial and river conditions succeeded each other in the passage

area more than once during the Cimmerian and Kuialnykian

times.

The strip of Cimmerian marine muds along the Galaţi

Passage (Mokriak et al. 2008) indicates a relatively high

sea-level stand and establishment of a connection between

the Dacian and Black Sea basins at the beginning of

the Cimmerian. Subsequently, a marked regional base-level

fall (Matoshko et al. 2009) very likely led to overflow of

the Dacian Basin into the Black Sea during the first half of

the Cimmerian (Fig. 10a). This generated the observed inci-

sion in the area of the Reni–Izmail Trough, which was fol-

lowed by deposition of the Near-Danube Suite in the incised

canyon. The presence of the brackish molluscs in the lowest

muds of this suite (Gozhik & Chirca 1973) may indicate re-

establishment of a narrow, 10–20 km wide strait. The early

Cimmerian corresponds to the Bosphorian (Late Pontian) of

the Dacian Basin in the latest timescales (Krijgsman et al. 2010;

Krijgsman & Piller 2012; Jorissen et al. 2018). Base-level

variations within the Bosphorian of the Dacian Basin are

poorly documented and it is not clear if the events we observe

in the Porat area are clearly expressed there.

Early Porat Substage: genesis of a large alluvial basin infill

The onset of deposition of the Porat Fm. probably came in

the Late Cimmerian (Dacian) (Fig. 2), when the sea level had

relatively stabilized (Fig. 10b). The major Early Porat Substage

of deposition lasted from the middle Zanclean (about 4.7 Ma)

to the late Gelasian (about 1.8–1.9 Ma). During the Dacian

Stage, the Dacian Basin was progressively infilled from

the west to the east, as indicated by the gradual replacement of

marine sedimentation by a fluvio–lacustrine environment

(Jipa & Olariu 2009; Olariu et al. 2018). The Dacian Basin

sub sequently became a fresh-water and predominantly fluvial

environment by the beginning of the Romanian (4.2 Ma;

Jipa & Olariu 2009; van Baak et al. 2015). Interfingering of

the Porat Fm. sands with frequent muds with fresh- and possi-

bly brackish water fauna (Gozhik & Chirca 1973) shows that

brackish-lacustrine conditions alternated with fluvial ones in

the Galati Passage. Van Baak et al. (2015), found that typical

lymnocardiide bivalve genera occur in a thin interval of

the Slanicul de Buzau section on the western margin of

the Focsani Depression, dated magnetostratigraphically at

2.95–3.20 Ma. This was interpreted to reflect a short-lived

incursion of the Black Sea during a base-level high stand

(Plescoi flooding event). This might possibly have a relation

with the time the Galati Passage developed as a lake.

While synsedimentary normal faulting would have helped

to keep the Galati Passage open, at some point the Porat infill

may have restricted it (Fig. 10c). This can have contributed to

the freshening of the Dacian Basin during the Romanian. It is

likely that part of the Porat infill was also directed towards

the Kuialnykian Black Sea Basin.

Late Porat Substage: appearance of Dolynske Valley and

arrival of the paleo-Danube to the Black Sea

The cutting of the Dolynske Valley and its subsequent infill

was a sharp event signifying the complete infilling of the Dacian

Lake and spillage of the Danube and its tributaries into the Black

Sea Basin. The specific mineral composition of the Dolynske

Mb. sand is very similar to that of sand from the present-day

Danube and its tributaries in the Western Dacian Basin

(de Leeuw et al. 2018). This means that the Dolynske Mb. pro-

vides the first tangible evidence for the presence of the Danube

in the Galati Passage, as earlier supposed by Konstantinova

(1967). We infer that the Danube and its main tributaries

the Siret and the Prut paved their final way through the area of

the former Dacian Lake and through the Galaţi Passage at

approximately 1.9–1.8 Ma BP (Fig. 10d), based on the age of

the youngest mammal fossils found in the Dolynske Mb. This

age is much younger than the 4 Ma age inferred by Olariu et

al. (2018), but in agreement with reconstructions by Andreescu

(2009) and Black Sea sediment provenance data (de Leeuw et

al. 2018). As was stated above, subsequent river entrenchment

during the Calabrian into deposits of the Dolynske Mb. meant

the end of the Porat Stage. It led to transformation of the pre-

vious coastal lowland into the elevated and erosionally dis-

sected plain with incised river valleys that it is today.

View on landscape–climatic features from sedimentary data

As indicated, there is a rich record with information on

the regional climate during deposition of the Porat Fm.

However, this record is very contradictory, which, as we will

show, is related to the nature of the sedimentary environment.

It is noteworthy that the lower part of the Porat Fm. includes

remains of large mammals belonging to many different bio-

topes (Ali-Zade et al. 1972): the riverside (beavers, water pigs,

elephants Deinotherium, large cats, hippopotamus); the forest

108

MATOSHKO, MATOSHKO and DE LEEUW

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

(bears, tapirs, macaques, monkeys Dolichopithecus, masto-

dons Borsoni, squirrels, deer Cervus unicolor, roe, lynxes);

the forest-steppe (mastodons of Anancus family, rhinoceros,

hipparions, bush-antlered deer, hares) and the arid steppe

(camels, antelopes, hyenas, corsac foxes, pikas).

The early diagenetic features of the Porat Fm. (reddish

weathering crust and soils with carbonate nodules) that were

noted before (e.g., Rengarten & Konstantinova 1965; Sinegub

1969; Hubca 1982) and observed during the course of this

study confirm that the region had a semi-arid climate (e.g.,

Alonso-Zarza 2003).

Today, the Porat domain is a dry steppe dis-

sected by narrow strips with river floodplains.

Therefore, the stable combination of forests

and steppes in the Plio cene–Lower Pleis-

tocene requires an expla nation, which we

think is evident in the sedimentary data.

The dynamic Porat infill provided surface

water and maintained a relatively high

groundwater level for the growth of woody

vegetation and other riverside- and forest spe-

cies in and close to the braided river belt.

Ample precipitation of moisture in the Car-

pathian Mountains provided much runoff for

the Porat and Siret rivers, which probably had

a very regular and distinct high-mean water

regime.

The Porat domain was thus a vast azonal

landscape on a river plain surrounded by a dry

steppe zone with a generally semi-arid cli-

mate. The warm and humid conditions of

the Cimmerian may have shifted the land-

scape zones southwards. Faunal elements

characteristic of different zones nevertheless

continued to coexist in Gelasian–Calabrian

times, which are frequently considered by

most researchers to have been a colder and

drier period in this area. The above interpreta-

tion from the point of view of the fluvial

process does not pretend to explain all the con-

tradictory questions in the evolution of

the biota, but speaks of the need for its inclu-

sion in further paleogeographic studies.

Conclusion

The focus of our attention was the Pliocene

to Calabrian Porat Fm., which occurs on

the borders of Romania, Moldova and Ukraine,

along the NE margin of the Dacian Basin.

The Porat Fm. is interpreted as a large, north–

south directed, wandering sandy alluvial

basin infill with a generally very uniform

aggradational structure, consisting of cyclic

channel–floodplain units deposited by rela-

tively powerful braided rivers. The sediment

supply for the Porat infill came from rewor-

king of the late Miocene Balta Fm. and

the overlying Pontian shallow marine strata

in addition to the significant contribution

Fig. 10. Fluvial sedimentary system evolution during the Porat Stage in comparison with

the Late Stage. Some maps of the Porat Stage are accompanied by rose diagrams of

the paleo current direction based on cross-bedding dip determinations. DB — Dacian

Basin, PF — Porat Fan, V — Porat–Prut, Siret and Danube valleys, FD — Focşani

Depression (basin depocenter); rivers: S. — Siret, B. — Bârlad, P. — Prut, Ya. — Yalpukh,

D. — Danube. Details see in the text.

109

THE PLIO–PLEISTOCENE FLUVIAL SYSTEM OF THE PALEO-DANUBE

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

from the Carpathians. The Porat basin infill generally overlies

shallow marine strata deposited during the Early Pontian

transgression with a disconformity, while it rests on early

Cimmerian strata along its southern margin. In this southern

strip, a Dolynske Mb. is distinguished, with a different alluvial

facies and sediment composition from the rest of the Porat

Fm., and representing the infill of a large west-east running

paleo-valley. We infer the Dolynske Mb. to have been depo-

sited by the Danube based on a similarity in heavy mineral

assemblages. The Danube, with its continental-scale drainage

basin stretching across the Alps, Dinarides and Carpathians,

has thus been supplying sediment to the Black Sea since at

least 1.9–1.8 Ma BP.

Aggradation of the Porat basin infill was in the first place

governed by down-warping of the Dacian Basin as the main

and stable tectonic background. The end of the Porat Stage

(the end of the Calabrian) is associated with final inversion of

crustal movements, leading to uplift and tilting of the whole

Porat area and whole basin of the Porat River. This inversion

corresponded to completion of the construction orogenic

phase in Easternmost Carpathians and the final establishment

of terrestrial conditions in the Dacian Basin, what resulted in

the onset of the last phase of the fluvial development conti-

nuing from the end of the Calabrian till today. The south-eastern

part of the Porat area was especially dynamic in the tectonic

respect. It was manifested in pronounced local block subsi-

dence along the New Trotus and St. George faults. This led to

formation of the Reni–Izmail Trough, which is paleogeo-

graphically known as the Galaţi Passage. This trough formed

the connecting gateway between the Black Sea and Dacian

Basin during the Pliocene–Gelasian–Calabrian. It allowed for

overflow of the Dacian Basin into the Black Sea during Black

Sea base-level lowstands and for brackish-marine ingressions

into the Dacian Basin during Black Sea base-level highstands.

The trough also manifests some of the most impressive river

entrenchments and the arrival of the Danube to the Porat area

around 1.9–1.8 Ma BP.

A stable delivery of water from the Carpathians allowed for

the existence of the Porat infill in a semi-arid climatic zone.

As a result the Porat area displayed an azonal forest–steppe

landscape with various faunal biotopes within coastal alluvial

lowland.

The relatively thin Porat Fm. is key for understanding not

only the development of its analogues within the north- western

coastal regions of Black Sea (such as Kuialnykian, Dniester

Liman and Roksolany formations), but also for correlation of

various events and phenomena in the Pliocene–Quaternary of

the residual Eastern Paratethys basins. Together with the Balta

Fm. it contains valuable information for further in-depth study

of the Carpathian foreland and its tectonic–sedimentary–relief

evolution.

Acknowledgements: We thank our colleagues and friends

from Ukraine, the UK and Moldova: I. Nicoara, S. Vincent,

V. Monastyretskii, and R. Spitsa, for friendly company in

the field and for retrieving some vital information; M. Stoica

for determinations of microfauna. We would like to sincerely

thank handling editor Michal Šujan, reviewer Dan Valentin

Palcu and an anonymous reviewer for their valuable comments

and suggestions that have significantly improved the quality

of the manuscript. The “GEOEXPERT” LLc (Ukraine) is

acknowledged for financial support. This work is the result of

our project entitled: “Sedimentary and surface evolution of

the marginal parts of platforms at the contact with active

tectonic belts”.

References

Ali-Zade A.A., Alizade K.A., Aleskarov D.A., Asadulaev E.A.,

Buleishvili D.A., Vekua A.K., Konstantinova N.A., Lebedeva N.A.,

Negadaev-Nikonov K.N., Nikiforova K.V., Pevzner M.A.,

Hubca A.N., Tchepalyga A.L. & Cherniakhovskii A.G. 1972:

Excursion Guidbook (Moldavia, Georgia, Azerbaijan). Moscow,

1–120 (in Russian).

Agustí J., Cabrera L., Garcés M., Krijgsman W., Oms O. & Parés J.M.

2001: A calibrated mammal scale for the Neogene of Western

Europe. State of the art. Earth-Sci. Rev. 52 , 247–260.

Alexandrova L.P. 1989: About detailed stratigraphic division of

Middle Pliocene alluvial deposits of Southern Moldavia (by

means of small mammals). Bulletin of Commission on Quater-

nary Period research 58, 97–104 (in Russian).

Alonso-Zarza A.M. 2003: Palaeoenvironmental significance of palus-

trine carbonates and calcretes in the geological record. Earth-

Sci. Rev. 60, 261–298.

Andreescu I. 2009: Magnetostratigraphy of the Upper Neogene depo-

sits in the eastern Dacian Basin: an overview. Muzeul Olteniei

Craiova. Oltenia. Studii și comunicări. Științele Naturii XXV,

329–336.

Andreescu I., Codrea V., Lubenescu V., Munteanu T., Petculescu A.,

Ştiucǎ E. & Terzea E. 2013: New developments in the Upper

Pliocene–Pleistocene stratigraphic units of the Dacian Basin

(Eastern Paratethys), Romania. Quaternary International 284,

15–29.

Bala A., Radulian N. & Popescu E. 2003: Earthquakes distribution

and their focal mechanism in correlation with the active tectonic

zones of Romania. J. Geodyn. 36, 129–145.

Bell C.J. & Bever G.S. 2006: Description and significance of the

Microtus (Rodentia: Arvicolinae) from the type Irvington fauna,

Alameda County, California. J. Vertebr. Palaeontol. 26, 2,

371–380.

Bilinkis G.M. 1992: Geodynamics of the south-western part of the

East European platform during the epoch of its morphogenesis.

Shtiintsa, Kishinev, 1–182 (in Russian).

Bilinkis, G.M., Bobrinskii V.M., Burgelia N.K., Dubinovskii V.L.,

Zheru M.I., Peres F.S., Ropot V.M., Smirnova V.A., Stratulat G.V.

& Shafranskii V.N. 1976: Pliocene bentonites of Moldaviia and

their practical importance. Shtiintsa, Kishinev, 1–211 (in Rus-

sian).

Bridge J.S. 2003: Rivers and Floodplains. Blackwell, London, 1–504.

Bridge J. & Demicco R. 2008: Earth Surface Processes, Landforms

and Sediment Deposits. Cambridge University Press, New York,

1–832.

Bukatchuk P.D., Gozhik P.F. & Bilinkis G.M. 1983: On correlation of

alluvial deposits of Dniester, Prut and Lower Danube. In: Zelenin

I.V. (Ed.): Geology of Quaternary deposits of Moldavia. Shtiintsa,

Kishinev, 35–82 (in Russian).

Cherednichenko V.G., Bukatchuk V.P. & Pokatilov 1985: Map of

pre-Quaternary formations. 1:1,000,000. Ministry of Geology of

the USSR, VSEGEI, Leningrad.

110

MATOSHKO, MATOSHKO and DE LEEUW

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

Chirca V.G. 1969: On Lower Antropogene subaqueous deposits in

the Lower Prut and Danube. In: Bondarchuk V.G. (Ed.): Mate-

rials on Quartenary Period of Ukraine. Naukova Dumka, Kiev,

176–197 (in Russian).

Čermák S. 2010: The Late Miocene and Pliocene Ochotoninae

(Lagomorpha, Mammalia) of Europe — the present state of

knowledge. In: Nowakowski D. (Ed.): Morphology and Syste-

matics of Fossil Vertebrates. DN Publisher, Wrocław, 9–28.

Clauzon G., Suc J.P., Popescu S.M., Marunteanu M., Rubino J.L.,

Marinescu F. & Melinte M.C. 2005: Influence of Mediterranean

sea-level changes on the Dacic Basin (Eastern Paratethys) during

the late Neogene: The Mediterranean Lago Mare facies deci-

phered: Basin Research 17, 437–462.

De Leeuw A., Morton A., van Baak C.G.C. & Vincent S.J. 2018:

Timing of arrival of the Danube to the Black Sea: Provenance of

sediments from DSDP site 380/380A. Terra Nova 30, 114–124.

Dodonov A.E., Sadchikova T.A., Tesakov A.S., Titov V.V. & Trubikhin

V.M. 2005: Issues of stratigraphy of Pliocene–Quaternary

formations of the north-western coast of the Black Sea. In:

Matishov G.G. (Ed.): Problems of Palaeontology and archeo-

logy of South of Russia and neighboring areas. “CVVR” Pub-

lishers, Rostov-na-Donu, 27–28 (in Russian).

Enciu P. & Dumitrică C. 2008: Professor Liviu Ionesi’s contribution

to the Palaeogeographical evolution of the Eastern Carpathians

Foreland knowledge. Acta Palaeontologica Romaniae 6, 97–103

(in Romanian).

Erbajeva M.A. & Shushpanov K.I. 1988: Pliocene pikas from Moldova.

Vestn. Zool. 4, 55–60 (in Russian).

Fejfar O., Heinrich W.-D., Kordos L. & Maul L.C. 2011: Microtoid

cricetids and the early history of arvicolids (Mammalia, Roden-

tia). Palaeontologia Electronica 14, 3, 27A. https://www.uv.es/

pe/2011_3/6_fejfar/index.html (accessed 3 June 2018).

Fielitz W. & Seghedi I. 2005: Late Miocene-Quaternary volcanism,

tectonic and drainage system evolution in the East Carpathians,

Romania. Tectonophysics 410, 111–136.

Forsten A. 1996. Climate and the evolution of Equus (Perissodactyla,

Equidae) in the Plio-Pleistocene of Eurasia. Acta Zoologica

Cracoviensia 39, 1, 161–166.

Ghenea C. 1968: Study of the Pliocene deposits between the Prut Val-

ley and the Bârlad Valley. [Studiul depositelor Pliocene dintre

valea Prutului şi valea Bîrladului.] Technical and Economic

Studies, Series J Stratigraphy 6, Bucharest, 1–137 (in Romanian

and French).

Ghenea C. 1997: The Pliocene–Pleistocene boundary in Romania.

In: Van Couvering J.A. (Ed.): The Pleistocene boundary and

the beginning of the Quaternary. Cambridge University press,

216–220.

Ghenea C. & Ghenea A. (Eds.). 1967: Romania. Geological map,

1:200,000, Birlad, L-35-XVI. Institute of Geology, Bucharest (in

Romanian).

Gozhik P.F. 2006: Freshwater mollusks of the Late Cenozoic of the

South of the Eastern Europe. Part 1. Geografica, Delta Ukraina,

Kiev, 1–280 (in Russian).

Gozhik P.F. (Ed.) 2012: Sratigraphic Code of Ukraine. 2-nd edition.

Kyiv, 1–66 (in Ukrainian).

Gozhik P.F. & Chirca V.G. 1973: New data on Pliocene deposits of

the lower reaches of the Prut and Danube rivers and problems

of their correlation. In: Bondarchuk V.G. (Ed.): On lower boun-

dary of the Quaternary. Naukova Dumka, Kiev, 66–72 (in

Russian).

Gradstein F.M., Ogg J.G., Schmitz M.D. & Ogg G.M. (Eds.)

2012: The Geological Time Scale 2012. Elsevier, Amsterdam,

1–1144.

Gromov I.M. & Polyakov I.Ya. 1977: Fauna of the USSR: Mammals.

III (8): Voles (Microtinae). Nauka, Leningrad, 1–248 (in

Russian).

Hubca A.N. 1982: Characteristics of the Carbalia Beds and their area

of occurrence. In: Adamenko O.M. (Ed.): Problems of the Antro-

pogene of Moldaviia. Shtiintsa, Kishinev, 36–75 (in Russian).

Hubca A.N., Tretiak A.N. & Volok Z.Eu. 1983: Stratigraphical posi-

tion of Carbalia Beds (by Palaeomagnetic data). In: Zelenin I.V.

(Ed.): Geology of the Quaternary deposits of Moldavia. Shtiintsa,

Kishinev, 82–89 (in Russian).

Ionesi L., Ionesi B., Roşca V., Lungu A. & Ionesi V. 2005: Middle and

Upper Sarmatian of the Moldovian Platform. Editura Academiei

Române, Bucharest, 1–558 (in Romanian).

Jipa D.C. & Olariu C. 2009: Dacian Basin. Depositional architecture

and sedimentary history of a Paratethys Sea. Geo-Eco-Marina,

Special Publication 3, 1–264.

Jorissen E.L., de Leeuw A., van Baak Ch..G.C., Mandic O., Stoica

M., Abels H.A. & Krijgsman W. 2018: Sedimentary architecture

and depositional controls of a Pliocene river-dominated delta in

the semi-isolated Dacian Basin. Sediment. Geol. 368, 1–23.

Kent D.V. 1999: Orbital tuning of geomagnetic polarity time-scales.

Philos. Trans. R. Soc. London Ser. A 357, 1995–2007.

Kholodov V.N., Golubovskaya E.V. & Nedumov R.I. 2014: Origin

and Prospects of the Cimmerian Iron Ore Basin in Ukraine and

Russia. Lithology and Minerals 5, 383–405.

Konstantinova N.A. 1967: Antropogene of the Southern Moldaviia

and South-Western Ukraine. Transactions of the Geological

Institute 173, 1–138 (in Russian).

Krijgsman W. & Piller W. 2012: Central and Eastern Paratethys

Regional Stages. In: Gradstein F.M., Ogg J.G., Schmitz M. &

Ogg, G.. The Geologic Time Scale 2012. Elsevier, Oxford,

935–937.

Krijgsman W., Stoica M., Vasiliev I. & Popov V.V. 2010: Rise and fall

of the Paratethys Sea during the Messinian Salinity Crisis. Earth

Planet. Sci. Lett. 290, 183–191.

KrijgsmanW., Tesakov A., Yanina T., Lazareva S., Danukalova G.,

Van Baak C.G.C., Agustí J., Alçiçek M.C, Aliyeva E., Bista, D.,

Bruch A., Büyükmeriç Y., Bukhsianidze M., Flecker R., Frolov

P., Hoyle T.M., Jorissen E.L., Kirscher U., Koriche S.A.,.

Kroonenberg S.B, Lordkipanidze D., Oms O., Rausch L.,

Singarayer J., Stoica M., van de Velder S., Titov V.V., Wesselinghr

F.P. 2019: Quaternary time scales for the Pontocaspian domain:

Interbasinal connectivity and faunal evolution. Earth-Sci. Rev.

188, 1–40.

Leever K., Matenco L., Bertotti G., Cloetingh S. & Drijkoningen G.

2006: Late orogenic vertical movements in the Carpathian Bend

Zone — seismic constraints on the transition zone from orogen

to foredeep. Basin Research 18, 521–545.

Lister A.M. & van Essen H. 2003: Mammuthus rumanus (ȘTE-

FĂNESCU), the earliest mammoth in Europe. In: Advances in

Vertebrate Palaeontology “Hell to Panla”, Part II. Romanian

Academy, Speleological Institute “Emil Racovita”, Bucharest,

47–53.

Matenco L., Bertotti G., Leever K., Cloetingh S., Schmid S.M. &

Tărăpoancă M. 2007: Large-scale deformation in a locked colli-

sional boundary: Interplay between subsidence and uplift, intra-

plate stress, and inherited lithospheric structure in the late stage

of the SE Carpathians evolution. Tectonics 26, 4, 1–29, TC4011.

Matoshko A.A., Matoshko A.V., de Leeuw A. & Stoica M. 2016:

Facies analysis of the Balta Formation: Evidence for a large late

Miocene fluvio-deltaic system in the East Carpathian Foreland.

Sedimentary Geology, 343, 165–189.

Matoshko A.V., Gozhik P.F. & Danukalova G. 2004: Key Late Ceno-

zoic fluvial archives of eastern Europe: the Dniester, Dnieper,

Don and Volga. Proceedings of the Geologists’ Association 115,

141–173.

Matoshko A.V., Gozhik P.F. & Semenenko V.N. 2009: Late Cenozoic

fluvial development within the Sea of Azov and Black Sea coas-

tal plains. Global Planet. Change 68, 270–287.

111

THE PLIO–PLEISTOCENE FLUVIAL SYSTEM OF THE PALEO-DANUBE

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

Maul L. & Markova A.K. 2007: Similarity and regional differences in

Quaternary arvicolid evolution in Central and Eastern Europe.

Quaternary International 160, 81–99.

Miall A.D. 2000: Principles of Sedimentary Basin Analysis. Springer-

Verlag, Berlin, Heidelberg, New York, 1–616.

Miall A.D. 2006: The geology of fluvial deposits. Sedimentary

Facies, Basin Analysis, and Petroleum Geology. Springer-

Verlag, Berlin, Heidelberg, New York, 1–582.

Mokriak I.M., Shvets L.K., Dobosh O.P. & Popov S.V. 2008:

L-35-XXIII (Izmail), L-35-XXIX (Tulcha). In: Velikanov V.Ya.

(Ed.): State geological map of Ukraine. 1:200,000. Kyiv (in

Ukrainian).

Munteanu I., Matenco L., Dinu C. & Cloetingh S. 2012: Effects of

large sea

‐level variations in connected basins: the Dacian–Black

Sea system of the Eastern Paratethys. Basin Research 24, 5,

583–597.

Muratov M.V. 1964: Cimmerian palaeogeography of the Middle Plio-

cene in area of the Black Sea — Caspian Sea basin. Lithology

and Minerals, 4, 3–20 (in Russian).

Necea D., Fieletz W. & Matenco L. 2005: Late Pliocene–Quaternary

tectonics in the frontal part of the SE Carpathians: Insights from

tectonic geomorphology. Tectonophysics 410, 137–156.

Negadaev-Nikonov K.N., Lukashev V.K., Hubca A.N. & Anoshko

Ya.I. 1980: Palaeontological and geochemical indicators of the

Quaternary environment in south-western part of the USSR.

Shtiintsa, Kishinev, 1–115.

Nichols G.J. & Fisher J.A. 2007: Processes, facies and architecture of

fluvial distributary system deposits. Sediment. Geol. 195, 75–90.

Nikiforova K.V., Alexandrova L.P., Trubikhin V.M. & Tchepalyga

A.L. 1986: Correlation of Pliocene–Eopleistocene deposits of

the South of the European part of USSR and SRR. In: Continen-

tal Upper Pliocene of the Black Sea — Caspian Sea region.

Rotaprint GIN AN USSR, Moscow, 5–17 (in Russian).

Olariu C., Krezsek C. & Jipa C. 2018: The Danube River inception:

Evidence for a 4 Ma continental-scale river born from segmen-

ted ParaTethys basins. Terra Nova 30, 1, 63–71.

Paleobiology Database 2018: Fossilworks. Macquarie University.

http://fossilworks.org. (accessed 6 June 2018)

Palatnaia N.N. 1991: Geological structure of the bottom deposits

of the Sasyk Liman. In: Shelkoplias V.N. (Ed.): Antropogene

(Quaternary) formations of Ukraine. Institute of Geological

Sciences of Academy of Sciences of Ukraine, Kiev, 126–130

(in Russian).

Palcu D.V., Golovina L.A., Vernyhorova Y.V., Popov S.V. &

Krijgsman W. 2017: Middle Miocene paleoenvironmental crises

in Central Eurasia caused by changes in marine gateway confi-

guration. Global Planet. Change 158, 57–71.

Popescu S-M., Dalesme F., Jouannic G., Escarguel G., Head M.J.,

Melinte

‐Dobrinescu M.C., Sütö‐Szentai M., Bakrac K., Clauzon

G. & Suc J-P. 2009: Galeacysta etrusca complex: Dinoflagellate

cyst marker of paratethyan influxes to the Mediterranean Sea

before and after the peak of the Messinian salinity crisis.

Palynology 33, 2, 105–134.

Popov S.V., Rögl F., Rozanov A.Y., Steininger F.F., Shcherba I.G. &

Kovač M. (Eds.) 2004: Lithological–Palaeogeographic maps of

Paratethys. 10 maps Late Eocene to Pliocene. E. Schweizer-

bart’sche Verlagsbuchhandlung.

Radulescu C. & Samson P-M. 1996: Pliocene and Early Pleistocene

arvicolids (Rodentia, Mammalia) of the Dacic Basin, Romania.

Acta Zoologica Cracovensia 39, 1, 401–406.

Radulescu C. & Samson P-M. 2001: Biochronology and evolution of

the Early Pliocene to the Early Pleistocene mammalian faunas of

Romania. Boll. Soc. Palaeontologica Italiana 40, 2, 285–291.

Radulescu C., Samson P-M., Petculescu A. & Stiucă E. 2003 : Plio-

cene large mammals of Romania. Coloquios de Palaeontología

1, 549–558.

Rengarten N.V. & Konstantinova N.A. 1965. The role of the facies–

mineralogic analysis in reconstruction of the Antropogene cli-

mate (on example of the Southern Moldaviia and South-Western

Ukraine). Transactions of the Geological Institute 137. 1–138

(in Russian).

Rögl F. 1998: Palaeogeographic considerations for Mediterranean

and Paratethys seaways (Oligocene to Miocene). Annalen des

Naturhistorischen Museums in Wien 99, 279–310.

Roşca V.Kh. 1969: Miocene. In: Polev P.V. (Ed.): Geology of the

USSR, XLV, Moldavskaia SSR. Geologic description and

minerals. Nedra, Moscow, 137–171 (in Russian).

Rosgen D.L. 1994: A classification of natural rivers. Catena 22, 169–199.

Ross D.A. 1978:. Summary of results of Black Sea drilling. In: Ross

D. A. & Neprochnov Y. P. (Eds.): Initial Reports of the Deep Sea

Drilling Project. U.S. Government Printing Office, Washington

DC, 1149–1178.

Rudkevich M.Ya. 1955: Some peculiarities of the Neogene history in

western part of the Fore-Black Sea Depression. Bulletin of Mos-

cow Society of Naturalists. Geological Branch XXX, 1, 1–20.

Rümke C.G. 1985: A review of fossil and recent Desmaninae

(Talpidae, Insectivora). Utrecht MicroPalaeontological Bulle-

tins 4, 1–241.

Sadchikova T.A., Alexandrova L.P., Trubikhin V.M. & Tchepalyga

A.N. 1983: Genetic conditions, palaeontological and palaeo-

magnetic characteristics of ancient fluviatile deposits of the

Southern Moldaviia (Luchesty Section) In: Upper Pliocene of

the Black and Caspian seas region. Rotaprint, GIN AN SSSR,

Moscow, 50–58 (in Russian).

Sanders C.A.E., Andriessenan P.A.M. & Cloetingh A.P.L. 1999: Life

cycle of the East Carpathian orogen: Erosion history of a doubly

vergent critical wedge assessed by fission track thermochronol-

ogy. J. Geophys. Res. 104, B12, 29,095–29,112.

Saulea E., Ghenea C., Bandrabur T. & Ghenea A. (Eds.) 1967: Roma-

nia. Geological map, 1:200,000. Focșani, L-35-XXII-XXIII.

Institute of Geology, Bucharest (in Romanian).

Shevchenko A.I. 1965: Key sites of the Pliocene and Lower Antropo-

gene of the south-western part of the Russian Plain. In: Strati-

graphic importance of Antropogene fauna of small mammals.

Nauka, Moscow, 7–59 (in Russian).

Shushpanov A.I. 1977: Fauna of the small mammals of the site Chish-

michioi in Moldavia. In: Yatsko I.Ya. (Ed.): Fauna complexes of

Cenozoic flora of the areas adjoined to the Black Sea. Shtiintsa,

Kishinev, 32–35 (in Russian).

Shushpanov A.I. 1980: Fauna of the small mammals of the Middle

and Upper Pliocene of the South-West of Moldavia. Proceedings

of the AN MSSR, series: biological and chemical sciences 5,

234–237 (in Russian).

Sincov I.F. 1897: Descriptions of some Neogene fossils found in Bes-

sarabiia. Notes of the Novorossiia Society of Naturalists 21, 1,

1–123 (in Russian).

Sinegub V.V. 1969: Pliocene. In: Polev P.V. (Ed.): Geology of the

USSR, XLV, Moldavskaia SSR. Geologic description and mine-

rals. Nedra, Moscow, 179–195 (in Russian).

Snel E., Mărunţeanu M., Macaleţ R., Meulenkamp J.E. & van Vugt N.

2006: Late Miocene to Early Pliocene chronostratigraphic

framework for the Dacic Basin, Romania. Palaeogeogr. Palaeo-

climatol. Palaeoecol. 238, 107–124.

Svetlitskaya T.V. 1994: Landscape and climate of the south-western

Russian Plain in the Pliocene. In: Thompson R.S. (Ed.): Pliocene

Terrestrial Environments and Data/Model Comparisons. U.S.

Geol. Survey Open-File Report 94-023, Version 1.0, http://pubs.

usgs. gov/of/1994/of94-023/text.txt (accessed 17 September 2017).

Tchepalyga A.L. 1967: Quaternary freshwater mollusks of the

southern regions of the Russian Plain and their stratigraphical

importance. Transactions of the Geological Institute, Geologi-

cal Series 166, 1–222 (in Russian).

112

MATOSHKO, MATOSHKO and DE LEEUW

GEOLOGICA CARPATHICA

, 2019, 70, 2, 91–112

Tesakov A.S. 2003: Early evolutionary stages of pitymyoid mimo-

myine voles (Pitymimomys, Arvicolinae, Cricetidae) from the

Early Villanyian of Eastern Europe. Coloquios de Palaeon-

tología 1, 659–668.

Tesakov A.S. 2004: Biostratigraphy of Middle Pliocene–

Eopleistocene of Eastern Europe (Based on Small

Mammals). Trans

actions of GIN RAN 554, 1–247

(in Russian).

Thomas R.G., Smith D.G., Wood J.M., Visser J., Calverley-Range

E.A. & Koster E.H. 1987: Inclined heterolithic stratification.

Terminology, description, interpretation and significance. Sedi-

ment. Geol. 53, 123–179.

Van Baak Ch.G.C., Mandic O., Iuliana L., Stoica M. & Krijgsman W.

2015: The Slanicul de Buzau section, a unit stratotype for the

Romanian stage of the Dacian Basin (Plio–Pleistocene, Eastern

Paratethys). Palaeogeogr. Palaeoclimatol. Palaeoecol. 440,

594–613.

Vangengeim E.A., Pevzner M.A. & Tesakov A.S. 1995: Chronologi-

cal relationships of Pliocene deposits in fluviatile plains between

Prut and Southern Bukh Rivers. Stratigraphy and Geological

Correlation 3, 1, 54–64.

Vangengeim E.A. & Tesakov A.S. 2008: Design principles of bio-

chronologic scales by means of mammals of the Pliocene and

Pleistocene. Problem state. Bulletin of Commission on Quater-

nary Period Research 68, 59–69 (in Russian).

Vasiliev I., Krijgsman W., Langereis C.G., Panaiotu C.E., Matenco L.

& Bertotti G. 2004: Towards an astrochronological framework

for the eastern Paratethys Mio-Pliocene sedimentary sequenses

of the Focsani basin (Romania). Earth Planet. Scie. Lett. 227,

231–247.

Vislobokova I. & Tesakov A. 2013: Early and Middle Pleistocene of

Northern Eurasia. In: Elias S.A. (Ed.): The Encyclopedia of

Quaternary Science, 4. Elsevier, Amsterdam, 605–614.

Wong H.K., Panin N., Dinu C., Georgescu P. & Rahn C. 1994:

Morphology and post-Chaudian (Late Pleistocene) evolution of

the submarine Danube Fan Complex: Terra Nova 6, 5, 502–511.

Zheru M.I. 1978: Clayey formations of Moldavia. Shtiintsa, Kishinev,

1–233 (in Russian).

Site name

East

longitude

North

latitude

Number

in Fig. 1

Albota de Jos

28.4707

45.9441

Andrușul de Sus

28.2575

46.0076

BăIăbăneşti 1

27.7305

46.0915

BăIăbăneşti 2

27.7331

46.0908

Bereşti

27.8904

46.1100

Bolgrad 1

28.6560

45.6758

Bolgrad 3

28.6500

45.6570

Borceag

28.4871

46.0651

Budăi 1

28.4634

45.8353

Budăi 2

28.5035

45.8110

Cahul 2

28.2357

45.8811

Chioselia

28.4124

46.1403

Chioselia Mare

28.4636

46.1030

Ciobalaccia 1

28.2986

46.1679

Ciobalaccia 2

28.2752

46.1520

Cislita–Prut

28.1723

45.5350

Cişmichioi 1,2

28.3892

45.5435

Cîșla

28.3337

46.1352

Cociulia 1

28.3972

46.3632

Colibași-Brînza

28.1741

45.6907

Colibași 1

28.1849

45.6970

Colibași 2

28.2088

45.7334

Crăciun

28.3593

46.2290

Dermengi

28.4125

45.8736

Dolynske 1

28.3088

45.4848

Dolynske 2

28.3143

45.4722

Dolynske 3

28.3151

45.4541

Dolynske 4

28.3265

45.4526

Etulia Noua 6

28.4472

45.5405

Etulia Noua 2

28.4352

45.5198

Flocoasa

28.2721

46.1369

Foltești

28.0583

45.7241

Frumusica

28.4691

46.0807

Gavanoasa

28.7744

45.3961

Giurgiuleşti 1

28.1994

45.4944

Giurgiuleşti 2

28.1829

45.4837

Site name

East

longitude

North

latitude

Number

in Fig. 1

Hîrtop-Balabanu

28.5181

45.9310

Huluboaia-Doina

28.3178

46.0819

Kotlovyna

28.5868

46.5188

Lărguța

28.3019

46.2823

Ljdileni 1

28.0328

45.6559

Ljdileni 2

28.0468

45.6380

Luceşti

28.3186

45.9911

Lymanske 1

28.4124

45.4196

Malusteni 1

27.9178

46.1908

Manta

28.2155

45.7963

Manzatesti

27.8767

46.1658

Moscovei

28.3012

45.9227

Musaitu

28.5035

45.8110

Pelinei

28.3256

45.8179

Slobozia Oancea

28.1061

45.8876

Spicoasa

28.3785

46.0445

Tartaul de Salcie

28.4213

45.9717

Tătăreşti 1

28.3511

46.0122

Tătăreşti 2

28.3537

46.0390

Topolyne

28.6511

45.6111

Trifestii Noi

28.3563

45.9577

Tuluceşti

28.0418

45.5610

Tutcani

27.9474

46.1730

Ursoaia

28.3223

45.8567

Valeni - Slobozia Mare 1

28.1666

45.6075

Valeni - Slobozia Mare 2

28.1696

45.6118

Valeni - Slobozia Mare 3

28.1921

45.5987

Valeni 1

28.1731

45.6590

Valeni 2

28.1766

45.6217

Vanatori

28.0361

45.5302

Vynogradivka1

28.5667

45.7107

Vladimirovca

28.3567

45.7887

Vulcănești 1

28.4463

45.6846

Vulcănești 2

28.3680

45.6756

Vulcănești 3

28.3664

45.6806

Appendix

Location of cited sites.